Campaigns | Digital Watch Observatory

DW Weekly #223 – AI race heats: The US AI Action Plan, China’s push for a global AI cooperation organisation, and the EU’s regulatory response

25 July – 1 August 2025

Book, Comics, Publication, Adult, Male, Man, Person, Face, Head, Clothing, Coat, Hat, James Montgomery Flagg

Dear readers,

Over the past week, the White House has launched a sweeping AI initiative through its new publication Winning the Race: America’s AI Action Plan, an ambitious strategy to dominate global AI leadership by promoting open-source technology and streamlining regulatory frameworks. America’s ‘open-source gambit’, analysed in detail by Dr Jovan Kurbalija in Diplo’s blog, signals a significant shift in digital policy, intending to democratise AI innovation to outpace competitors, particularly China.

Supporting this bold direction, major tech giants have endorsed former President Trump’s AI deregulation plans, despite widespread public concerns regarding potential societal impacts. Trump’s policies notably include an explicit push for ‘anti-woke’ AI frameworks within US government contracts, raising contentious debates about the ideological neutrality and ethical implications of AI systems in governance.

In parallel, China has responded with its own global AI governance plan, proposing the establishment of an international AI cooperation organisation to enhance worldwide coordination and standard-setting. Thus, it is not hard to conclude that there is an escalating AI governance competition between the two technological superpowers, each advocating distinctly different visions for the future of global AI development.

On the multilateral stage, the UN’s Economic and Social Council (ECOSOC) adopted a resolution: ‘Assessment of the progress made in the implementation of and follow-up to the outcomes of the World Summit on the Information Society’, through the Commission on Science and Technology for Development (CSTD), reaffirming commitments to implement the outcomes of the World Summit on the Information Society (WSIS).

Corporate strategies have also reflected these geopolitical undercurrents. Samsung Electronics has announced a landmark $16.5 billion chip manufacturing deal with Tesla, generating optimism about Samsung’s capability to revive its semiconductor foundry business. Yet, execution risks remain substantial, prompting Samsung’s Chairman Jay Y. Lee to promptly travel to Washington to solidify bilateral trade relations and secure the company’s position amid potential trade tensions.

Similarly, Nvidia has placed a strategic order for 300,000 chipsets from Taiwanese giant TSMC, driven by robust Chinese demand and shifting US trade policies.

Meanwhile, the EU has intensified regulatory scrutiny, accusing e-commerce platform Temu of failing mandatory Digital Services Act (DSA) checks, citing serious risks related to counterfeit and unsafe goods.

In the USA, similar scrutiny arose as Senator Maggie Hassan urged Elon Musk to take decisive action against Southeast Asian criminal groups using Starlink services to defraud American citizens.

Finally, the EU’s landmark AI Act commenced its implementation phase this week, despite considerable pushback from tech firms concerned about regulatory compliance burdens.

Diplo Blog – The open-source gambit: How America plans to outpace AI rivals by democratising tech

Jovan Kurbalija

On 23 July, the US unveiled an AI Action Plan featuring 103 recommendations focused on winning the AI race against China. Key themes include promoting open-source AI to establish global standards, reducing regulations to support tech firms, and emphasising national security. The plan addresses labour displacement, AI biases, and cybersecurity threats, advocating for reskilling workers and maintaining tech leadership through private sector flexibility. Additionally, it aims to align US allies within an AI framework while expressing scepticism toward multilateral regulations. Overall, the plan positions open-source AI as a strategic asset amid geopolitical competition. Read the full blog!

For the main updates, reflections and events, consult the RADAR, the READING CORNER and the UPCOMING EVENTS section below.

Join us as we connect the dots, from daily updates to main weekly developments, to bring you a clear, engaging monthly snapshot of worldwide digital trends.

DW Team

RADAR

Highlights from the week of 25 July – 1 August 2025

Free VPN use surges in UK after online safety law

But worries rise as many free VPNs exploit users or carry hidden malware

australia flag is depicted on the screen with the program code

Australia reverses its stance and restricts YouTube for children under 16

From December, YouTube must block accounts for Australians under 16 or face massive fines.

aeroflot cyberattack silent crow belarus cyber partisans ukraine conflict

Aeroflot cyberattack cripples Russian flights in major breach

Belarusian and Ukrainian hackers claim responsibility for strategic cyber sabotage of Aeroflot.

2025%2F07%2Fbeautiful silhouette port machinery sunset

NATO highlights cyber vulnerabilities in European ports

A NATO policy brief warns that civilian ports across Europe face increasing cyber threats from state-linked actors and calls for updated maritime strategies to strengthen cybersecurity and civil–military coordination.

Italy investigates Meta over AI integration in WhatsApp

AGCM says Meta may have harmed competition by embedding AI features into WhatsApp.

Google backs EU AI Code but warns against slowing innovation

The EU AI Code could add €1.4 trillion to Europe’s economy, Google says.

US stablecoin supply jumps $4 billion after new crypto laws

Tether and Circle dominate the fiat-backed stablecoin market, now valued at over $227 billion combined.

EU clears Microsoft deal after privacy changes

Brussels updates Microsoft terms to curb risky data transfers

Teachers and students warn: AI is eroding engagement

AI use in schools is weakening the connection between students and teachers by permitting students to bypass genuine effort through shortcuts.

Employee depression rising amid AI burnout and job insecurity

Use of AI surveillance, including monitoring software, intensifies burnout, micromanagement feelings, and disengagement.

Big companies grapple with AI’s legal, security, and reputational threats

A majority of Fortune 500 companies now mention AI in their annual reports as a risk factor instead of citing its benefits.

man using laptop night workplace top view

Crypto hacks soar in 2025 as security gaps widen

The platforms lost more than $3.1 billion in the first half of 2025, with AI-powered hacks and phishing scams leading the surge.

AI reshaping the US labour market

AI jobs now span marketing, finance, and HR—not just tech.

jojickajoja27 quantum computing 91572e2a e7d0 40d4 b6e5 12dc1bca48c6 11zon

AI and quantum tech reshape global business

Google and Microsoft lead investment in advanced AI and quantum infrastructure.

View all updates

READING CORNER

BLOG featured image 2025 The open source gambit

The open-source gambit: How America plans to outpace AI rivals by democratising tech – Diplo

ChatGPT Image Jul 28 2025 at 10 13 23 PM

How are we being tracked online?

Tracking technologies shape our online experience in often invisible ways, yet profoundly impactful, raising important questions about transparency, control, and accountability in the digital age.

18 – 25 July 2025

machine, Wheel, Person, Bulldozer, Face, Head

Dear readers,

This week, President Donald Trump officially signed the GENIUS Act into law, marking a historic step in establishing a legal framework for stablecoins in the US. The act, passed with bipartisan support on 18 July, introduces the first rules for the $250 billion stablecoin market.

Additionally, Trump Media and Technology Group, backed by $2.5 billion in funding, has acquired around $2 billion worth of Bitcoin as part of an investment plan announced earlier this year.

The EU and Japan formally reaffirmed their long-standing EU‑Japan Digital Partnership during the third Digital Partnership Council in Tokyo.

Meta has refused to endorse the EU’s new voluntary Code of Practice for general-purpose AI, citing legal overreach and risks to innovation. In addition, Meta has urged the Australian government to harmonise privacy regulations with international standards, warning that stricter local laws could hamper AI development.

OpenAI has signed a strategic partnership with the UK government aimed at strengthening AI security research and exploring national infrastructure investment.

ChatGPT now receives over 2.5 billion daily prompts, with more than 330 million originating from users in the US.

A growing number of US teens rely on AI for daily decision‑making and emotional support, with chatbots such as ChatGPT, Character.AI and Replika.

The UK government has unveiled a new proposal to strengthen its response to ransomware threats by requiring victims to report breaches, enabling law enforcement to disrupt cybercriminal operations more effectively.

A sophisticated new ransomware threat, dubbed GLOBAL GROUP, has emerged on cybercrime forums, meticulously designed to target systems across Windows, Linux, and macOS with cross-platform precision.

The Dutch Public Prosecution Service (OM) had confirmed a significant cyberattack that forced it to disconnect from the internet, following warnings of a potential vulnerability.

Iran’s Minister of Communications has revealed the country’s digital economy shrank by 30% in just one month, losing around $170 million due to internet restrictions imposed during its recent 12-day conflict with Israel.

North Korea is dispatching AI researchers, interns and students to countries such as Russia in an effort to strengthen its domestic tech sector, according to a report by NK News.

OpenAI has officially signed a $30 billion annual agreement with Oracle to expand data centre capacity in the US.Amazon is shutting down its AI research lab in Shanghai, marking another step in its gradual withdrawal from China.

For the main updates, reflections and events, consult the RADAR, the READING CORNER and the UPCOMING EVENTS section below.

Join us as we connect the dots, from daily updates to main weekly developments, to bring you a clear, engaging monthly snapshot of worldwide digital trends.

DW Team

RADAR

Highlights from the week of 18 – 25 July 2025

Top Cybersecurity Courses to Apply For in 2022

Crypto crime surges to record levels in 2025

Chainalysis warns 2025 may see double the number of physical attacks on crypto holders compared to previous years.

Japan breaks internet speed record with 1.02 Pbps

Power demands reshape future of data centres

New AI workloads are pushing server power demands past 100kW per rack, stressing existing infrastructure.

Policy gaps widen Europe’s 5G divide

Northern and southern Europe now lead the EU in 5G coverage, thanks to timely spectrum rollout

Amazon cuts hundreds of AWS jobs amid AI shift

AWS sales and profits are rising, yet Amazon is cutting jobs within the division, aiming to balance investment with efficiency through AI-driven task automation instead of expansion.

Louis Vuitton Hong Kong Cyberattack data breach

Louis Vuitton Australia confirms customer data breach after cyberattack

Following the cyberattack, Louis Vuitton advised clients to watch for suspicious emails and calls, even though no financial details were accessed.

Co op cyberattack Khoury Haq data breach

Co-op confirms massive data breach as retail cyberattacks surge

Retail cyberattacks continue with breach at Co-op.

Salt Typhoon targets routers in sweeping campaign

Operators report only client-side devices affected—but the campaign reflects telecom as a growing espionage target.

Italy challenges tech giants over VAT on user data

The dispute questions how digital services and personal data are valued under EU law.

Spotify Blaze Foley AI music Syntax Error

Spotify under fire for AI-generated songs on memorial artist pages

AI controversy follows similar streaming incidents involving fake artist profiles.

fraud scam phishing caution deception concept

Surge in UK corporate data leaks fuels fraud fears

Most corporate data leaks now involve sensitive financial and HR records, warns major cybersecurity study.

Android malware MediaPlayer JavaScript Telegram

Android malware infects millions of devices globally

BadBox 2.0 spreads through cheap gadgets, triggering adware, ransomware, and wider attacks.

Bitcoin’s security under quantum threat

Around 7 million BTC could be vulnerable to quantum-based attacks, a cybersecurity expert warns.

Teikoku Databank cyberattacks Japan SMEs

Historic UK KNP transport firm collapses after ransomware attack

Hackers from Akira encrypted vital systems and demanded about £5 million, forcing a shutdown.

teens in circle holding smart mobile phones

Teens struggle to spot misinformation despite daily social media use

New research reveals teens wrongly believe misinformation only affects world news or scams, leaving them vulnerable to more subtle forms of fake content.

View all updates

READING CORNER

AI chatbots Grok Tay Microsoft Prometheus Sydney LaMDA Lamoine Elon Musk

Not just bugs: What rogue chatbots reveal about the state of AI

When AI chatbots go rogue, the fallout reveals more about human choices than machine intent—or technical limits.

Rwanda Congo Peace diplomacy blog featured image

Testing the limits of peace: What the DRC-Rwanda deal reveals about mediation today – Diplo

The DRC–Rwanda peace accord of June 2025 marks a rare moment of successful peace diplomacy, aiming to end decades of conflict in the Great Lakes region through troop withdrawals, security coordination, and economic cooperation.

Are democracies drifting toward psephocracy and technocracy? – Diplo

When politicians avoid hard choices, courts and experts step in to fill the gap. Power shifts away from voters to institutions that don’t answer to them. What does this mean for democracy’s future? Aldo Matteucci writes.

29 July 2025 – 31 July 2025

Diplo/GIP at Sixth World Conference of Speakers of Parliaments

Diplo/GIP at Sixth World Conference of Speakers of Parliaments On 29–31 July 2025, the Inter-Parliamentary Union (IPU), in collaboration with the United

11 – 18 July 2025

People, Person, Architecture, Building, Outdoors, Shelter, Countryside, Hut, Nature, Rural, Adult, Male, Man, Face, Head, Crowd, Audience, Octagón

Dear readers,

The past week has witnessed a surge in cyberattacks, underscoring the growing vulnerability of digital systems amid evolving threats.

Hackers stole $500,000 in cryptocurrency by deploying a malicious Cursor AI extension, exploiting AI supply chains and highlighting the risks of trusted tools turning into attack vectors.

The Salt Typhoon campaign intensified, compromising critical US infrastructure, including telecommunications and government networks, prompting the US House to pass the NTIA Cyber Leadership Bill to formalise agency coordination after severe telecom breaches.

In the UK, a cyberattack on Co-op exposed the personal data of 6.5 million members, prompting a public apology from its CEO, while in Azerbaijan, over 6,200 government workers fell victim to poor cyber hygiene, exposing public sector weaknesses.

Meta faced backlash after a small business lost thousands due to a breach. Android users, instead, were targeted with trojanized Telegram APKs exploiting the Janus vulnerability, alongside MacOS devices hit by a fake Termius app.

Google urged caution as Gmail’s AI tools faced new threats, signalling risks in automated systems.

This wave of incidents, driven by sophisticated tactics and geopolitical tensions, reveals a critical need for robust defences. However, the reactive nature of legislative and corporate responses raises doubts about their ability to keep pace with this escalating threat landscape.

Quantum computing

The week behind us also marked a significant leap forward in quantum computing, spotlighting Europe’s ambitious push to reclaim global influence. The EU announced a bold investment strategy to bolster quantum technology to counterbalance the US’s private-sector dominance and China’s state-driven efforts. The initiative includes the development of quantum computers using glass and light, with European researchers unveiling a scalable quantum light factory chip that promises enhanced computational power. Complementing this, a groundbreaking project to create the world’s first single-photon universal quantum system, slated for 2026, showcases Europe’s focus on practical quantum applications. These advancements position Europe as a key player amidst a global race, though the narrative of regained influence is tempered by challenges in matching the US’s private funding and China’s centralised resources, suggesting a competitive yet uneven playing field.

EU’s digital policy developments of the week

The EU confirmed the rollout of the AI Act, effective August 2025, and released the General-Purpose AI (GPAI) Code of Practice to guide compliance for leading AI providers, aiming to balance innovation with safety. However, legal uncertainty looms for non-signatories, as the code’s voluntary nature may burden companies proving compliance through alternative means, raising questions about enforcement equity.

Meanwhile, Apple faces accusations of stifling browser competition on iOS, prompting scrutiny under the Digital Markets Act (DMA), while a s u rge in AI-generated fake news tests the EU Digital Services Act’s ability to curb misinformation, exposing potential regulatory gaps. The EU’s launch of the EuroSky satellite constellation, announced this week, seeks to reclaim digital sovereignty by enhancing secure connectivity, though its success hinges on funding and geopolitical cooperation.

Meta encountered fresh EU backlash for DMA non-compliance, particularly over data practices, intensifying calls for stricter oversight.

The Netherlands urged the EU to reduce reliance on US cloud providers, advocating for a Europe-first digital infrastructure. Yet, this shift risks fragmenting the market and increasing costs, challenging the narrative of seamless integration.

Chip export control side effects

Nvidia announced plans to resume AI chip sales to China following productive US-China talks, signalling a potential thaw in restrictions that had strained supply chains, though compliance with US export rules remains a focal point. Nvidia’s CEO, Jensen Huang, asserted that the Chinese military’s use of US chips is unlikely, aiming to ease concerns and justify the restart. Yet, this claim lacks independent verification, raising questions about oversight effectiveness.

Meanwhile, Malaysia enforced stricter trade controls on AI chips of US origin, aligning with US efforts to curb China’s access to advanced technology, which has disrupted regional semiconductor markets and prompted local firms to seek alternative suppliers.

For the main updates, reflections and events, consult the RADAR, the READING CORNER and the UPCOMING EVENTS section below.

Join us as we connect the dots, from daily updates to main weekly developments, to bring you a clear, engaging monthly snapshot of worldwide digital trends.

DW Team

RADAR

Highlights from the week of 11 – 18 July 2025

Vatican urges ethical AI development

Ethical AI must prioritise the common good over profit or efficiency.

Bitcoin surge sparks global crypto job boom

Crypto jobs are going mainstream as market value climbs past $3.8 trillion and top economies offer stronger policy support.

EXA to boost European connectivity with new fibre route and subsea cable

EXA Infrastructure has launched a major 1,200 km fibre route connecting key European hubs, featuring the first new North Sea subsea cable in 25 years.

trump pope AI image White House Truth Social

Trump unveils AI economy with $100 billion investment push

Pennsylvania to become a key player in Trump’s ‘AI economy’, as major energy and tech projects worth $100 billion are announced during Pittsburgh summit.

https hypebeast.com image 2025 06 20 meta and oakley unveil all new ai powered oakley meta hstn glasses 003

Meta opens audio lab to improve AI smart glasses

Cambridge facility is Meta’s largest engineering investment outside the US, signalling a push for leadership in immersive AI wearables.

Children turn to AI chatbots instead of real friends

Experts warn that children are using AI chatbots instead of seeking real friendships, raising concerns about emotional dependency and blurred lines between human and machine.

Defence AI Centre at heart of Korean strategy

Seoul’s bold push to merge cutting-edge AI with its defence industry is set to redefine how nations balance security, innovation, and ethical responsibility in the age of intelligent warfare.

El Salvador’s Bitcoin reserves surge past $760 million

President Bukele’s controversial Bitcoin policy is yielding long-term value, despite ongoing international criticism.

Trump launches $70 billion AI and energy investment plan

Renewables face rollback under Trump’s infrastructure blueprint.

Military AI and the void of accountability

Military AI is advancing faster than global rules can keep up, raising urgent questions about accountability, hidden biases, and the terrifying possibility of wars escalating beyond human control.

US moves to block Chinese tech in undersea Internet cables

The US plans to ban undersea cables using Chinese technology from connecting to its networks, citing rising cybersecurity and espionage risks.

Elon Musk’s xAI secures $2 billion from SpaceX

SpaceX backs Musk’s xAI venture with $2 billion, linking space technology and AI instead of keeping the two sectors separate.

zoom video header still landingspagina in het kort.jpg 1280x1280 q85 subsampling 2

AI tour retraces Anne Frank’s wartime route

Featuring 12 stops and archival animations, the self-guided route highlights Miep Gies and other rescuers, offering fresh insight into courageous wartime acts.

the pentagon in washington dc politics military

Pentagon awards AI contracts to xAI and others after Grok controversy

Elon Musk’s xAI has secured a $200 million Pentagon contract, raising ethics questions following recent scandals involving its Grok chatbot on social media.

View all updates

READING CORNER

The Grand Ethiopian Renaissance Dam BLOG featured image

The Grand Ethiopian Renaissance Dam as a test case for 21st-century energy diplomacy – Diplo

An in-depth look at the Grand Ethiopian Renaissance Dam (GERD) and the complex diplomatic conflict over water and energy between Ethiopia, Egypt, and Sudan.

Do diplomats have personal rights to protection under the VCDR that they can enforce? – Diplo

Do diplomats have personal rights under international law, or are all rights held by the sending state? Alan Franklin examines why diplomatic immunity often leaves individuals with ‘a right without a remedy’.

No judges, no appeals, no fairness: Wimbledon 2025 shows what happens when AI takes over

AI made the calls at Wimbledon — and sparked global debate.

UN OEWG concludes, paving the way for a permanent cybersecurity mechanism

The OEWG on ICT security has adopted its Final Report after intense negotiations on responsible state behaviour in cyberspace.

21 July 2025

July 2025 online courses | Diplo Academy

July 2025 online courses | Diplo Academy Diplo Academy is excited to announce the start of three online courses on 21 July 2025:

23 July 2025

Five years on: Achievements, failures, and the future of the UN Cyber Dialogue

Five years on: Achievements, failures, and the future of the UN Cyber Dialogue The one-hour session will bring together four discussants – two representatives

19 Jul 2025 – 25 Jul 2025

IETF 123

The 123th meeting of the Internet Engineering Task Force (IETF) will take place from 19 to 25 July 2025 in Madrid, Spain.

Rétrospective de juin 2025

Chères lectrices,
Cher lecteurs,

En juin, tous les regards étaient tournés vers le 20e Forum sur la gouvernance de l’Internet (IGF) 2025 à Lillestrøm, en Norvège, où DiploAI a fait sensation en tant que partenaire officiel chargé des comptes rendus, couvrant chaque session grâce à une transcription assistée par intelligence artificielle de pointe, et diffusant des analyses en temps réel via la plateforme dig.watch.

Dans un contexte mondial marqué par des débats sur la gouvernance de l’IA, la souveraineté numérique et la cybersécurité, le forum a ouvert la voie à un dialogue innovant. De la réduction de la fracture numérique à la lutte contre la désinformation liée à l’IA, les tendances observées en juin révèlent un monde à la croisée des chemins, où la technologie façonne la géopolitique, le commerce et les droits de l’Homme.

Rejoignez-nous pour découvrir les développements clés du mois, en mettant en lumière les moments marquants du FGI 2025 et le panorama plus large de la diplomatie numérique.

Quelques faits marquants du mois de juin 2025 dans le domaine du numérique :

Le FGI 2025 s’ouvre en Norvège : le forum a débuté le 23 juin et rassemble plus de 8 000 parties prenantes afin de définir la gouvernance de l’internet.
L’UE alloue 145,5 millions d’euros à la cybersécurité – ces fonds sont destinés à la protection des soins de santé et des infrastructures critiques.
Le bitcoin se maintient au-dessus des 100 000 dollars – la stabilité des cryptomonnaies témoigne de leur adoption croissante dans un contexte de changements réglementaires.
Le sommet de l’OTAN aborde les cybermenaces – des hacktivistes pro-russes ciblent les infrastructures lors du 76e sommet.
Le CDH examine les technologies et les droits de l’Homme – les rapports sur l’IA et l’enregistrement numérique des naissances ont été débattus le 26 juin.
Le Conseil de l’UIT adopte le budget 2026-2027 et met l’accent sur l’inclusion numérique et la diplomatie de l’IA à Genève.
WhatsApp est interdit sur les appareils de la Chambre des représentants des États-Unis – le service de Meta est restreint pour des raisons de sécurité.
Un ordinateur quantique en orbite grâce à SpaceX – une avancée photonique laisse entrevoir de futurs changements technologiques.
L’UE demande un report de la loi sur l’IA, l’industrie invoquant l’absence de cadres réglementaires et l’incertitude juridique.
Les pays du Sud poussent pour l’inclusion numérique – une session du FGI met l’accent sur la gouvernance pour les personnes vulnérables.

L’analyse et le travail de terrain de Diplo à une époque exceptionnelle
Dans un monde où l’histoire évolue à une vitesse effrénée, le véritable défi n’est pas seulement de suivre le rythme, mais d’en comprendre le sens. Chaque jour apporte son lot d’informations, mais la vision d’ensemble se perd souvent dans le vacarme ambiant. Comment les événements actuels façonnent-ils les tendances à long terme ? Quel est leur impact sur nous en tant qu’individus, communautés, entreprises et même en tant qu’humanité?

Chez Diplo, nous faisons le lien entre les actualités en temps réel et les analyses plus approfondies. Notre Digital Watch suit de près les développements quotidiens tout en les reliant aux tendances hebdomadaires, mensuelles et annuelles, comme illustré ci-dessous.
Disk

De la cybersécurité au commerce électronique en passant par la gouvernance numérique, nous suivons ces changements, des fluctuations quotidiennes aux pivots industriels à long terme.

Dans le numéro mensuel #101 de juin, vous pourrez suivre: TENDANCES en matière d’IA et de technologie | Développements à GENÈVE | Dig.Watch ANALYSE

Les tendances du mois : Juin 2025

Rétrospective du mois de juin : IGF et tendances numériques

Le mois de juin 2025 a été marqué par l’impact retentissant du 20e Forum sur la gouvernance de l’Internet (IGF) 2025, qui s’est tenu du 23 au 27 juin à Lillestrøm, en Norvège, au Nova Spektrum, le plus grand rassemblement des Nations unies jamais organisé dans ce pays. Avec plus de 4 000 participants en personne et en ligne, le forum, intitulé « Construire ensemble la gouvernance numérique », a été ouvert par Chengetai Masango, chef du secrétariat de l’IGF des Nations unies, et Karianne Oldernes Tung, ministre norvégienne de la Numérisation, qui ont souligné la nécessité d’une vision collaborative pour un internet ouvert et sécurisé dans un contexte de tensions mondiales croissantes. Les principales sessions de la première journée ont porté sur le rôle de l’IA dans l’humanité et la fracture numérique, alors que 2,6 milliards de personnes ne sont toujours pas connectées à Internet. Le Dr Jovan Kurbalija a quant à lui appelé à un renouvellement du mandat de l’IGF et à un financement innovant avant le SMSI+20, afin de jeter les bases d’une gouvernance inclusive.

Le dialogue s’est intensifié lors de la deuxième journée avec la présentation du système indien Aadhaar (80 millions d’utilisateurs quotidiens) et du système brésilien PIX (5,7 milliards de dollars d’économies). Cependant, les défis en matière de financement sont considérables, et le Dr Kurbalija et Sorina Teleanu ont plaidé en faveur d’une coopération en matière d’IA et d’un engagement parlementaire. Les troisième et quatrième jours ont été consacrés à la diversité culturelle dans la gouvernance de l’IA et au projet d’écologie du savoir, qui a permis de mettre à disposition 19 ans de données du FGI pour les ODD, tandis que la dernière journée a été consacrée à la fragmentation de l’internet au titre de l’article 29C du Pacte mondial pour le numérique (GDC), Marilia Maciel appelant à la recherche économique et Gbenga Sesan soulignant les disparités entre les utilisateurs. Parmi les principales conclusions, citons la promotion de cadres mesurables, la collaboration multipartite soutenue et le rôle central de DiploAI dans la fourniture de rapports en temps réel basés sur l’IA via dig.watch, soulignant ainsi le mois de juin comme un mois crucial pour la politique numérique.

Émergence de cadres de gouvernance de l’IA

Le mois de juin a été marqué par une avancée décisive dans le domaine des cadres de gouvernance de l’IA, qui s’est concrétisée par la convergence des efforts institutionnels, régionaux et de la société civile présentés lors de l’événement dig.watch. Lors du FGI 2025 à Lillestrøm, des sessions telles que l’Examen de haut niveau de la gouvernance de l’IA ont mis en évidence une concentration préoccupante des capacités de calcul dans un petit groupe de pays, appelant à une démocratisation de l’accès aux GPU, à des audits algorithmiques et à une évaluation rigoureuse des modèles afin de lutter contre les taux d’hallucination de 26 % et les résultats biaisés. Parallèlement, le Forum mondial de l’UNESCO sur l’éthique de l’IA a réaffirmé son engagement en faveur de la recommandation de 2021, mettant l’accent sur les droits de l’Homme, la sensibilité au genre, l’alignement des multiples parties prenantes et les résultats en matière de durabilité. Des ateliers du FGI tels que « AI Innovation Responsible Development Ethical Imperatives » ont souligné l’unité éthique mondiale et mis l’accent sur des systèmes publics inclusifs, de l’agriculture à l’éducation, afin de garantir que l’IA serve des objectifs sociaux larges.

Par ailleurs, le Forum ouvert n° 33 du FGI, animé par l’Administration chinoise du cyberespace, a fait écho à la nécessité d’une coopération équitable, insistant sur l’importance de normes réglementaires communes et de garanties d’intérêt public. Des sessions telles que « L’IA à la croisée des chemins entre souveraineté et durabilité » ont approfondi les débats sur la justice environnementale et l’autonomie stratégique, en examinant l’utilisation non durable de l’eau par les centres de données et les vulnérabilités liées aux ressources des infrastructures nationales d’IA.

Au-delà des sommets mondiaux, dig.watch a également souligné la révision par l’AGNU d’un projet de dialogue sur la gouvernance de l’IA le 4 juin, considérée comme une avancée décisive vers une diplomatie scientifique inclusive. Le texte révisé prévoit une gestion équilibrée et la présentation d’un rapport annuel à l’Assemblée générale. Les plateformes de renforcement des capacités en matière d’IA, telles que le forum de l’UNESCO qui s’est tenu les 4 et 5 juin à Paris et les ateliers du Conseil de l’UIT, soulignent la nécessité d’investir dans la formation à la gouvernance et la préparation au déploiement de l’IA dans le secteur public.

La souveraineté à l’ère numérique – Gouvernance des données et obligations légales

Au cœur de l’Europe, le ministère danois de la Numérisation a lancé une migration à l’échelle gouvernementale des produits Microsoft vers LibreOffice et Linux, faisant écho aux efforts parallèles déployés à Copenhague et Aarhus pour renforcer l’autonomie nationale et le contrôle des données sur les plateformes basées aux États-Unis. Non loin derrière, l’État allemand du Schleswig-Holstein a officiellement abandonné Microsoft Teams et Office, invoquant des vulnérabilités géopolitiques et des inégalités en matière de licences, tout en redirigeant le stockage des données vers des infrastructures gérées par l’Allemagne. Ces mesures synchronisées interviennent alors que Lyon, troisième ville de France, est également en train de migrer ses systèmes municipaux vers des solutions open source telles que Nextcloud et PostgreSQL, révélant une vague de souveraineté municipale plus large qui marque le pivotement collectif de l’Europe dans le domaine des technologies de l’information du secteur public.

Parallèlement, l’UE a lancé DNS4EU le 9 juin, établissant un système de résolution de noms de domaine centré sur la confidentialité et conforme au RGPD, géré par un consortium paneuropéen afin de libérer la région des services DNS dominés par les États-Unis et de renforcer l’infrastructure continentale. Au Royaume-Uni, les leaders technologiques réévaluent leur dépendance vis-à-vis des fournisseurs de cloud américains, la plupart des professionnels de l’informatique soutenant le « rapatriement du cloud » afin de récupérer la souveraineté des données et de préserver l’autonomie stratégique dans le contexte du CLOUD Act.

Lors du FGI 2025 à Lillestrøm, les réseaux communautaires et les plateformes open source sont apparus comme des éléments centraux des discussions sur la souveraineté, mis en évidence lors de quatre sessions quotidiennes qui ont mis en avant les infrastructures gérées par la base afin d’équilibrer les systèmes centralisés et contrôlés par les entreprises.

Cette tendance autarcique s’est reflétée à l’échelle mondiale dans des sessions telles que « L’IA à la croisée des chemins entre souveraineté et durabilité », où les intervenants ont mis en évidence les interdépendances écologiques, telles que les centres de données qui consomment de l’eau, et ont plaidé en faveur d’un « modèle de solidarité numérique » donnant la priorité aux urgences communes des pays du Sud plutôt qu’à la fragmentation motivée par les ressources.

La gouvernance des données est étroitement liée à la justice, comme l’a souligné la troisième journée du FGI 2025, qui a mis l’accent sur la diversité culturelle dans l’IA, avec le projet d’écologie du savoir au service des ODD. « La société civile fait pression pour les droits numériques et la justice dans l’examen SMSI+20 » a exigé la responsabilité, et le rapport technique du CDH (A/HRC/59/32) a abordé l’impact de l’IA sur les droits de l’Homme.

Tensions géopolitiques dans le domaine cybernétique

Le mois de juin a été marqué par une escalade des tensions géopolitiques dans le domaine cyber, avec une convergence d’événements qui a révélé un paysage mondial où les infrastructures numériques sont désormais au cœur des conflits stratégiques.

L’adoption par le Japon, au début du mois, de la loi sur la cyberdéfense a symbolisé un changement de paradigme : la souveraineté cyber nationale fait désormais partie de la politique de défense, ce qui permet de prendre des mesures préventives contre les serveurs hostiles avant qu’ils ne puissent frapper des infrastructures critiques.

De l’autre côté de l’Atlantique, le sommet de l’OTAN a pris un tournant décisif : les États membres ont non seulement convenu d’augmenter leurs budgets de défense à 5 % du PIB, mais ils ont également reconnu officiellement le cyberespionnage et le sabotage par des groupes liés à la Russie comme des menaces de premier plan, ce qui a donné lieu à des exercices conjoints visant à renforcer la cyber-résilience des réseaux alliés.

Dans le mème temps, la recrudescence des opérations hacktivistes liées aux tensions entre l’Iran et Israël a signalé un développement imprévisible des zones de conflit cybernétique. Un résumé du dig.watch sur la cybercriminalité a signalé une forte augmentation des attaques contre les compagnies aériennes et les systèmes gouvernementaux, en particulier dans le contexte de la crise israélo-iranienne.

Au-delà des conflits entre États, les cybermenaces se sont intensifiées au niveau sociétal, des aéroports aux établissements de santé, avec de nouvelles campagnes de logiciels malveillants exploitant des plateformes omniprésentes telles que Zoom pour compromettre les actifs cryptographiques.

Évolution réglementaire des cryptomonnaies

Le mois de juin a été marqué par une évolution rapide de la réglementation et de l’adoption des cryptomonnaies : avancée des mesures réglementaires concrètes, évolutions du marché et acceptation par le grand public. En Europe, le très attendu cadre MiCA s’est rapproché de sa mise en œuvre d’ici 2025, signalant l’intention de l’UE de réglementer les stablecoins, de renforcer la transparence et de protéger les consommateurs. En Corée, une intense activité politique et législative a accompagné un nouveau projet de loi habilitant les entités commerciales à émettre des stablecoins.

De l’autre côté de l’Atlantique, les États-Unis ont connu des développements spectaculaires : le bitcoin a dépassé les 100 000 dollars, soutenu par des signaux politiques et un intérêt institutionnel plus large. Dans le mème temps, les projets de l’Agence fédérale de financement du logement visant à autoriser la détention de cryptomonnaies dans les actifs de réserve hypothécaire, encouragés par la dynamique présidentielle, ont mis en évidence la curiosité croissante des régulateurs au cœur du système financier.

Les régulateurs ont également renforcé la protection des consommateurs et l’application de la loi : au Royaume-Uni, Barclays a bloqué les transactions cryptographiques sur les cartes de crédit en raison des risques pour les consommateurs, tandis qu’à Singapour, la MAS a exigé des entreprises cryptographiques locales qu’elles cessent leurs services de jetons numériques à l’étranger d’ici le 30 juin, sous peine d’amendes et de peines de prison, renforçant ainsi les protocoles KYC et AML.

À ces tendances s’ajoute l’impact considérable de l’informatique quantique sur la sécurité des cryptomonnaies : les experts ont averti que des avancées telles que la puce Majorana de Microsoft pourraient permettre de pirater le Bitcoin d’ici cinq ans, soulignant un défi existentiel imminent pour les blockchains actuelles et appelant à investir de toute urgence dans des systèmes résistants à l’informatique quantique.

Commerce et guerre des puces IA : quand les contrôles à l’export deviennent des remparts pour la cybersécurité et la vie privée

Taïwan a imposé en urgence des contrôles à l’exportation visant TSMC, à la suite de la vente involontaire de “chiplets” à Huawei — un incident qui a poussé Taipei à renforcer la surveillance de tous les transferts de semi-conducteurs avancés. Parallèlement, le mégaprojet Stargate en matière d’IA lancé par les Émirats arabes unis, confronté aux préoccupations de Washington en raison de sa proximité avec des régimes sanctionnés, illustre comment les investissements non occidentaux dans l’intelligence artificielle sont désormais conditionnés au respect de paramètres géopolitiques sensibles.

Réduire la fracture numérique et garantir les droits numériques

Le FGI 2025 à Lillestrøm a débuté par des statistiques qui donnent à réfléchir : 2,6 milliards de personnes restent hors ligne, un chiffre alarmant qui redéfinit la connectivité comme un impératif démocratique et non plus seulement un objectif technique. Des sessions telles que « Closing Digital Divides by Universal Access & Acceptance » (Réduire la fracture numérique grâce à l’accès universel et à l’acceptation) ont réuni des intervenants du Canada, du Kenya et du Pakistan afin d’explorer les obstacles socio-économiques et linguistiques qui entravent l’inclusion (turn0search0). Le cadre ROAMX approuvé par l’UNESCO a été mis en avant pour son rôle dans l’évaluation des politiques numériques sous l’angle des droits, de l’ouverture, de l’accessibilité, de la participation multipartite et des considérations liées au genre et à la durabilité.

Les droits numériques ont occupé une place centrale, portés par la société civile lors du FGI 2025, qui a appelé à davantage de redevabilité dans le processus du SMSI+20. Le Conseil des droits de l’homme (CDH) a, de son côté, mis l’accent sur la protection des groupes marginalisés. Une session intitulée « La société civile s’oppose aux dérives des lois sur la cybersécurité » a dénoncé les abus étatiques, tandis que la déclaration de la Freedom Online Coalition sur l’IA a réaffirmé l’importance des droits fondamentaux — malgré des lacunes persistantes concernant l’usage des logiciels espions dans les pays du Sud.

La diversité linguistique a également occupé le devant de la scène, un groupe du FGI soulignant que la véritable connectivité nécessite une interface dans la langue maternelle, un appel soutenu par des experts de l’ICANN, d’Unicode et des autorités réglementaires de toutes les régions

Sur le plan politique, la publication en juin du « document de référence du SMSI+20 » a mis l’accent sur l’accès équitable, la justice des données, la cybersécurité et l’IA inclusive comme piliers de la coopération numérique mondiale.

Le canton de Genève met en œuvre un droit constitutionnel à l’intégrité numérique, se plongeant dans les normes opérationnelles en matière de protection des données et de lutte contre la surveillance, un banc d’essai expérimental à la frontière des droits numériques en Europe.

Parallèlement, une table ronde du FGI sur le thème « Click with Care » a regroupé les questions de la sécurité des enfants, de l’impact des algorithmes et des discours de haine dans un discours unifié sur les droits numériques, incitant les parlementaires à réclamer des lois sur la responsabilité des plateformes et une conception centrée sur l’humain.

Les tendances de juin 2025, des perspectives du FGI 2025 aux changements mondiaux, façonnent notre avenir numérique. Pour en savoir plus, rendez-vous sur dig.watch!

Juin 2025 à Genève

Water, Waterfront, City, Urban, Harbor, Pier, Metropolis, Architecture, Building, Cityscape, Outdoors, Boat, Transportation, Vehicle, High Rise, Windmill

En juin, Genève a réaffirmé sa position de centre névralgique mondial de la politique numérique, en proposant un riche programme d’événements, de sommets et d’initiatives multilatérales qui ont défini l’avenir de la gouvernance numérique.

Le mois a débuté avec le Giga School Connectivity Forum, qui s’est tenu le 5 juin au Campus Biotech et a réuni des acteurs de différents secteurs, notamment des représentants du gouvernement, du monde universitaire et des télécommunications, afin d’analyser les défis à relever pour parvenir à un accès universel à Internet dans les écoles d’ici 2030, en soulignant l’urgence de combler les fractures éducatives et numériques. Peu après, le rapport final de la réunion de haut niveau SMSI+20, qui s’est tenue le 10 juin à Genève et a réuni des diplomates, des responsables des Nations unies et des technologues afin de définir les priorités pour l’examen de l’Assemblée générale des Nations unies en décembre 2025, sur des questions telles que l’inclusion numérique, la cybersécurité et les cadres inclusifs pour l’IA, a consolidé le rôle de la ville dans l’élaboration des normes numériques mondiales.

À la mi-juin, la session du Conseil de l’UIT (17-27 juin) s’est réunie à nouveau au siège de l’UIT. L’atelier « AI in Action » et les discussions sur le Giga Connectivity Centre, lancé en partenariat avec l’UNICEF et l’Espagne au Campus Biotech, ont mis en évidence la manière dont l’IA et la connectivité sont intégrées dans l’agenda institutionnel de Genève. Parallèlement, le 9 juin, l’initiative DNS4EU a retenu l’attention. Bien qu’il s’agisse d’un projet distinct, son lancement souligne l’alignement de Genève sur les objectifs européens en matière de souveraineté des infrastructures.

Genève accueille également la 59e session du Conseil des droits de l’homme des Nations Unies, qui a lieu du 16 juin au 11 juillet. Outre les questions traditionnelles relatives aux droits, les délégués ont abordé les préoccupations liées à l’ère numérique : l’impact de l’IA sur la liberté d’expression, la surveillance et les droits des populations vulnérables. Bien qu’aucune résolution spécifique au numérique n’ait encore été présentée, la session a réaffirmé que le Conseil considère les droits numériques dans le cadre plus large des droits de l’Homme, ouvrant ainsi la voie à un engagement plus profond sur la confidentialité en ligne, la désinformation et l’inclusion numérique dans les prochains examens mondiaux.

Du 19 au 30 juin, la Geneva Internet Law Summer School a accueilli des universitaires du monde entier pour débattre de questions allant de la protection des données à la neutralité du net, soulignant ainsi le statut de Genève en tant que laboratoire pour la politique juridique numérique. Parallèlement, l’organisation ARTICLE 19, basée à Genève, a animé une session marquante exigeant que les droits numériques et l’équité soient au cœur des résultats du SMSI+20, présentant l’inclusion numérique comme une question indissociable des droits humains.

Coup d’oeil : Les développements qui font des vagues en juin

Le mois de juin 2025 a été marqué par des développements notables en matière de gouvernance de l’IA, de cybersécurité et de politique numérique mondiale. Voici un aperçu de ce qui s’est passé au cours du mois dernier:

TECHNOLOGIE

Le fabricant chinois de puces Loongson a dévoilé de nouveaux processeurs pour serveurs qu’il prétend comparables aux processeurs Ice Lake 2021 d’Intel, marquant ainsi une avancée dans la quête d’autonomie technologique du pays.

L’Afrique est largement à la traîne dans la course mondiale au développement de l’IA, selon un nouveau rapport de l’université d’Oxford.

Une avancée révolutionnaire a eu lieu dans l’exploration spatiale. Le premier ordinateur quantique photonique au monde a été mis en orbite avec succès à bord de la mission Transporter 14 de SpaceX.

Le département américain de la Défense a attribué à OpenAI un contrat de 200 millions de dollars pour développer des prototypes d’outils d’IA générative à usage militaire.

Les récentes avancées en matière d’informatique quantique ont ravivé les craintes concernant la sécurité à long terme du Bitcoin (BTC).

Un dispositif quantique de petite taille, développé par des chercheurs de l’Université de Vienne, a surpassé les algorithmes avancés d’apprentissage automatique classique, y compris certains utilisés dans les systèmes d’IA actuels, en utilisant seulement deux photons et une puce en verre.

Orange Business et Toshiba Europe ont lancé à Paris le premier service commercial de réseau quantique sécurisé en France.

Des physiciens de l’université d’Oxford ont réalisé une première mondiale dans le domaine de l’informatique quantique en établissant un nouveau record de précision pour le fonctionnement d’un qubit unique.

Oxford Quantum Circuits (OQC) a dévoilé son projet de développer d’ici 2034 un ordinateur quantique résistant aux pannes de 50 000 qubits, en utilisant sa technologie supraconductrice « Dimon » exclusive.

Des scientifiques chinois ont créé le premier système au monde basé sur l’IA capable d’identifier les ogives nucléaires réelles parmi les leurres, marquant ainsi une étape importante dans la vérification du contrôle des armements.

GOUVERNANCE

La nouvelle Stratégie numérique internationale de l’UE pour 2025 (publiée le 5 juin 2025) s’éloigne de la diplomatie numérique de l’UE, fondée sur ses valeurs, pour adopter une approche davantage axée sur la géopolitique, la sécurité et la concurrence.

Le Sénat américain a adopté la loi GENIUS, premier projet de loi établissant un cadre fédéral pour la réglementation des stablecoins adossés au dollar.

Le Vietnam a officiellement légalisé les cryptoactifs dans le cadre d’une loi historique sur les technologies numériques adoptée par l’Assemblée nationale le 14 juin.

INFRASTRUCTURE

Les politiques de défense et de sécurité en constante évolution du gouvernement britannique visent à combler les lacunes juridiques révélées par les menaces modernes telles que les cyberattaques et le sabotage des câbles sous-marins.

Les projets de création d’un vaste centre de données sur l’IA aux Émirats arabes unis ont suscité des inquiétudes en matière de sécurité à Washington en raison des liens étroits que ce pays entretient avec la Chine.

JURIDIQUE

Le ministère danois de la Culture a présenté un projet de loi visant à protéger les images et les voix des citoyens dans le cadre de la législation nationale sur le droit d’auteur.

Le président américain Donald Trump a annoncé une prolongation de 90 jours accordée à ByteDance, la maison mère chinoise de TikTok, pour trouver un acquéreur américain, reportant ainsi l’interdiction nationale de la populaire application de partage de vidéos.

OpenAI tenterait de réduire le contrôle exclusif de Microsoft sur l’hébergement de ses modèles d’IA, ce qui témoigne de tensions croissantes entre les deux entreprises.

Un juge fédéral de New York a ordonné à l’Office of Personnel Management (OPM) des États-Unis de cesser de partager des données personnelles sensibles avec les agents du Department of Government Efficiency (DOGE).

La Commission européenne a infligé une amende de 329 millions d’euros à la société berlinoise Delivery Hero et à sa filiale espagnole Glovo pour avoir participé à ce qu’elle a décrit comme un cartel sur le marché de la livraison de repas en ligne.

ECONOMIE

Le directeur général d’OpenAI, Sam Altman, a révélé avoir eu une conversation avec le PDG de Microsoft, Satya Nadella, lundi afin de discuter de l’avenir de leur partenariat.

Le fondateur de SoftBank, Masayoshi Son, envisage ce qui pourrait devenir son projet le plus audacieux à ce jour : un parc industriel dédié à l’IA et à la robotique d’une valeur de 1 000 milliards de dollars en Arizona.

Le cours du Bitcoin s’est effondré à la mi-juin en raison de l’aggravation des tensions géopolitiques au Moyen-Orient.

OKX a étendu sa présence en Europe en lançant des bourses centralisées parfaitement conformes en Allemagne et en Pologne.

Taïwan a officiellement interdit l’exportation de puces et de micro-puces vers les entreprises chinoises Huawei et SMIC, rejoignant ainsi les États-Unis dans le renforcement des restrictions sur les transferts de semi-conducteurs de pointe.

Amazon va investir 20 milliards de dollars australiens pour développer son infrastructure de centres de données en Australie, en utilisant l’énergie solaire et éolienne à la place des sources d’énergie traditionnelles.

SÉCURITÉ

Une attaque par rançongiciel lancée contre l’ONG suisse Radix a entraîné le vol et la publication en ligne de données gouvernementales sensibles.

Le conflit entre l’Iran et Israël s’est désormais étendu au cyberespace, où des groupes de pirates informatiques rivaux lancent des séries d’attaques à caractère politique.

La société chinoise d’IA DeepSeek gagne du terrain sur les marchés mondiaux malgré les préoccupations croissantes en matière de sécurité nationale.

Le Centre national de cybersécurité (NCSC) a publié de nouvelles directives pour aider les organisations à se conformer aux exigences de la prochaine directive européenne sur la sécurité des réseaux et de l’information (NIS2).

Une coalition d’agences de cybersécurité, dont la NSA, le FBI et la CISA, a publié des directives communes pour aider les organisations à protéger leurs systèmes d’IA contre les nouvelles menaces en matière de sécurité des données.

Dans un geste audacieux soulignant les préoccupations croissantes concernant la souveraineté numérique, l’État allemand du Schleswig-Holstein rompt ses liens avec Microsoft.

Avant la décision officielle du gouvernement danois, les villes de Copenhague et d’Aarhus avaient déjà annoncé leur intention de réduire leur dépendance à l’égard des logiciels et des services cloud de Microsoft.

Le Conseil de l’UE et le Parlement européen sont parvenus à un accord politique visant à renforcer l’application transfrontalière du règlement général sur la protection des données (RGPD).

L’une des plus importantes fuites de données de connexion jamais enregistrées a été découverte, exposant plus de 16 milliards d’enregistrements provenant de services largement utilisés, tels que Facebook, Google, Telegram et GitHub.

Une cyberattaque viserait la banque iranienne Bank Sepah, qui aurait été perpétrée par le groupe de hacktivistes Predatory Sparrow.

Plus de 20 000 adresses IP et domaines liés à des logiciels malveillants destinés à voler des données ont été neutralisés lors de l’opération Secure, une répression coordonnée de lutte contre la cybercriminalité menée par INTERPOL entre janvier et avril 2025.

Le FBI a lancé une alerte concernant la résurgence de BADBOX 2.0, un programme malveillant dangereux qui infecte des millions d’appareils électroniques grand public dans le monde entier.

Le président américain Donald J. Trump a signé un nouveau décret visant à modifier les politiques fédérales existantes en matière de cybersécurité.

Le Parlement japonais a adopté une nouvelle loi autorisant des mesures actives de cyberdéfense, permettant aux autorités de surveiller légalement les données de communication en temps de paix et de neutraliser les serveurs étrangers en cas de cyberattaques.

DÉVELOPPEMENT

L’OTAN examine actuellement des suggestions visant à élargir la couverture des dépenses liées à la défense afin d’aider les États membres à atteindre l’objectif proposé de 5 % du PIB.

SOCIO-CULTUREL

La plateforme de réseaux sociaux X a mis à jour son contrat de licence pour les développeurs afin d’interdire l’utilisation de son contenu pour le développement de grands modèles linguistiques.

L’Autorité des services financiers du Royaume-Uni (FCA) a pris des mesures contre des influenceurs financiers non autorisés dans le cadre d’une opération internationale coordonnée, aboutissant à trois arrestations.

Wikipédia a suspendu un essai controversé de synthèses d’articles générées par l’IA suite à de violentes réactions de sa communauté de rédacteurs bénévoles.

TikTok a interdit à l’échelle mondiale le hashtag « SkinnyTok » après avoir subi des pressions du gouvernement français, qui accusait la plateforme de promouvoir des habitudes alimentaires néfastes chez les jeunes utilisateurs.

Pour plus d’informations sur la cybersécurité, les politiques numériques, la gouvernance de l’IA et d’autres sujets connexes, veuillez consulter le site diplomacy.edu.

dig.watch Analyses

Faire progresser la triade suisse de l’IA : l’entrepreneuriat zurichois, la gouvernance genevoise et la subsidiarité communautaire

La Suisse a la capacité de se démarquer dans la course mondiale à l’IA en combinant trois atouts : l’esprit d’entreprise innovant de Zurich, la gouvernance responsable de Genève et la subsidiarité communautaire.

La « Swiss AI Trinity » peut aller au-delà de la technologie en réinventant un contrat social national pour l’ère de l’IA, fermement ancré dans ses valeurs. La transformation de la Suisse en matière d’IA évolue sans cesse alors que le monde technologique est en pleine mutation. Tandis que les États-Unis incitent à une croissance technologique sans entraves et que la rivalité entre les États-Unis et la Chine s’intensifie, ressemblant à une course à l’armement dans le domaine de l’IA, l’UE et les nations du monde entier cherchent leurs propres stratégies. L’incertitude grandit parallèlement à l’inquiétude croissante concernant l’impact de l’IA, qu’il s’agisse de l’emploi, de l’économie, de l’éducation ou des médias.

Dans cet environnement en pleine mutation, la Suisse dispose d’une occasion unique de se forger une approche inédite en matière de développement de l’IA de pointe, ancrée dans la subsidiarité, l’apprentissage et les traditions nationales.

Le temps presse, mais le moment est propice pour la Suisse :

L’IA est banalisée : le retard pris par la Suisse dans le développement des LLM n’est plus un désavantage. De nouveaux LLM apparaissent chaque jour, les modèles open source facilitent la reconversion et les agents IA peuvent être créés en quelques minutes. Le succès dépend désormais des données, des connaissances et des utilisations de l’IA, et non plus uniquement du matériel ou des algorithmes. La clé réside dans le lien entre l’intelligence artificielle et l’intelligence humaine, domaine dans lequel la Suisse peut exceller grâce à son solide modèle d’éducation et d’apprentissage.

Recul de la réglementation de l’IA : après une course effrénée à la réglementation ces dernières années, une approche plus équilibrée privilégie désormais les risques à court terme (emplois, éducation) plutôt que les menaces existentielles à long terme. L’administration Trump a déjà ralenti la dynamique réglementaire aux États-Unis, tandis que l’UE réévalue certaines parties de sa loi sur l’IA concernant les modèles génératifs. La prudence de la Suisse en matière de réglementation est devenue un avantage. Une nouvelle réglementation de l’IA pourrait être adoptée à mesure que la technologie mûrit et que les véritables problèmes politiques à réglementer se cristallisent.

Dans ce contexte, la « triade suisse de l’IA » propose une stratégie en trois volets :

Zurich : un pôle d’innovation du secteur privé ;
Genève : un vivier pour la gouvernance mondiale et la normalisation ;
Les cantons, les villes et les communes suisses : le respect de la subsidiarité pour favoriser un développement inclusif et ascendant de l’IA.

Chaque pilier s’appuie sur les atouts existants tout en répondant au besoin urgent de repenser les modèles économiques, la gouvernance et les contrats sociaux à l’ère de l’IA.

Zurich : dynamiser les entreprises et l’innovation

Alors que les technologies de pointe semblent souvent concentrées dans d’immenses centres de données et des entreprises pesant plusieurs milliards de dollars, DeepSeek illustre parfaitement comment des avancées révolutionnaires peuvent voir le jour dans des laboratoires modestes et d’avant-garde. Zurich est idéalement placée pour tirer parti de ces deux approches.

Écosystème de classe mondiale : l’ETH Zurich, classée parmi les meilleures universités au monde, ainsi que les centres de R&D de Microsoft, Google et d’autres entreprises, fournissent les talents, l’excellence en matière de recherche et l’esprit d’entreprise nécessaires pour maintenir la Suisse à la pointe de l’IA.
Une portée mondiale et des perspectives diverses : les partenariats doivent s’étendre au-delà de la Silicon Valley. L’implication des acteurs technologiques chinois et indiens favorise la concurrence, stimule la créativité et atténue la dépendance excessive à l’égard des chaînes d’approvisionnement occidentales.
Des liens plus étroits entre le monde universitaire et l’industrie : une collaboration plus approfondie entre l’ETH et les leaders du secteur privé accélérera les initiatives. Les percées dans les domaines de la santé, des technologies climatiques et au-delà apparaissent lorsque la recherche de pointe rencontre les applications concrètes.

Zurich n’est donc pas seulement un centre financier mondial, c’est également un modèle en matière d’IA responsable et avant tout centrée sur l’humain.

Genève : forger une gouvernance et des normes mondiales

Alors que Zurich stimule l’innovation, Genève peut façonner une gouvernance mondiale équilibrée de l’IA, une course qui s’intensifie face aux initiatives concurrentes du Golfe et des pôles technologiques émergents. Genève doit agir de manière décisive :

Transformer la frénésie en actions concrètes : les pays ont besoin de toute urgence d’outils pragmatiques basés sur l’IA pour lutter contre les pandémies, les risques environnementaux et garantir une éducation équitable. Les organismes basés à Genève devraient donner la priorité à des solutions concrètes, afin de dissiper la perception de l’IA comme un concept abstrait et fantaisiste.
Intégrer l’IA : il est essentiel de considérer l’IA comme partie intégrante de la santé, du commerce et des droits du travail. À mesure que l’IA devient une nécessité universelle, les institutions de Genève doivent intégrer la politique numérique dans toutes les négociations mondiales importantes.
Moderniser les organisations internationales : la gestion obsolète et trop centralisée des organisations internationales manque de souplesse pour s’adapter à l’évolution rapide de l’IA. L’intégration de services de traduction IA, de rapports automatisés et d’outils similaires renforcera la transparence, l’efficacité et la collaboration.
Définir des normes là où les traités font défaut : en l’absence d’accords internationaux solides, les normes techniques garantissent l’interopérabilité. L’expertise de Genève en matière de normalisation lui permet d’être à la pointe dans les domaines de l’IA appliquée à la santé, du commerce numérique et des données environnementales.

En acceptant ce rôle, Genève peut s’engager à ce que les instances multilatérales orientent l’éthique de l’IA dans un monde en pleine mutation.

Communautés et cantons : l’inclusion par la subsidiarité

Le principal avantage de la Suisse en matière d’IA réside dans la subsidiarité, c’est-à-dire le principe de la prise de décision au niveau local. La répartition du développement de l’IA entre les cantons et les communes garantit que l’innovation répond aux besoins réels, tient compte des contextes locaux et ne laisse personne de côté. Les principales activités sont les suivantes :

Programme « AI for All » : offrir des subventions modestes aux citoyens et aux petites entreprises pour développer des agents IA, démocratiser l’accès aux outils et catalyser des solutions ancrées dans les besoins locaux.
Formation et apprentissage en IA : développer l’apprentissage de l’IA, en s’appuyant sur la longue tradition suisse de formation professionnelle ; intégrer l’enseignement de l’IA de l’école primaire à l’université.
Bibliothèques et laboratoires locaux d’IA : reconvertir les bibliothèques, les centres communautaires et les bureaux de poste en centres de connaissances sur l’IA. L’application de l’apprentissage automatique à des défis hyperlocaux – agriculture, tourisme, santé – peut donner aux communautés les moyens de devenir des créateurs actifs d’innovations en matière d’IA.

Un appel à l’action : la Suisse doit prendre les devants

La Suisse se trouve à une période charnière. La synergie de Zurich, Genève et ses cantons peut favoriser un avenir de l’IA à la pointe de la technologie, mais aussi équitable, transparent et profondément helvétique.

En adoptant cette approche « Triade IA » , la Suisse peut montrer au monde entier comment adopter les technologies de pointe sans sacrifier les valeurs sociétales.

Mesures concrètes pour aller de l’avant :

Développer les apprentissages en IA dans le cadre de la tradition professionnelle suisse.
Lancer un programme national de renforcement des capacités en IA pour les citoyens et les entreprises.
Adapter les programmes scolaires et universitaires afin de favoriser la créativité, à mesure que l’IA automatise des tâches telles que la rédaction de dissertations.
Réaffecter les bibliothèques et les bureaux de poste en centres de connaissances communautaires.
Donner la priorité aux principes de l’« IA pour tous » dans les projets publics, les marchés publics et les subventions à tous les niveaux gouvernementaux.

L’approche suisse « Triade IA » peut démocratiser l’IA, favoriser l’innovation locale et stimuler une croissance inclusive à partir de la base. Les outils, les talents et la tradition sont en place. Il est temps d’agir.

Pour plus d’informations sur ces sujets, veuillez consulter le site diplomacy.edu.

Vous souhaitez en savoir plus sur l’IA, les technologies et la diplomatie numérique ? Si oui, Interrogez notre Chatbot Diplo!

Analyse de l’héritage du GTCNL : réussites, échecs et travaux inachevés

Le GTCNL sur la cybersécurité (2019-2025) a façonné les débats mondiaux sur la sécurité numérique, mais a-t-il tenu ses promesses ? Des experts externes évaluent son impact durable, tandis que notre équipe, qui a suivi le processus depuis le début, analyse les étapes majeures et les occasions manquées. Ensemble, ces perspectives révèlent ce que l’avenir réserve à la gouvernance cybernétique dans un monde fragmenté.

Qu’est ce que le GTCNL ?

Les groupes de travail ouverts (GTCNL) sont un système présent au sein des Nations unies généralement considéré comme le plus ouvert. Cela signifie que tous les États membres et observateurs des Nations unies, les organisations intergouvernementales et les organisations non gouvernementales dotées d’un statut consultatif auprès du Conseil économique et social des Nations unies (ECOSOC) peuvent assister aux réunions publiques du groupe de travail. Cependant, les décisions sont prises par les États membres des Nations unies. Il existe différents GTCNL au sein des Nations unies. Nous nous intéressons ici à celui qui traite de la cybersécurité.

Que fait le GTCNL sur la cybersécurité ? En termes simples, il s’efforce de trouver un terrain d’entente sur ce qui est autorisé et ce qui ne l’est pas dans le cyberespace, et sur la manière de garantir le respect de ces règles. Dans le langage des Nations unies, le GTNCL du cyberespace a été chargé de « continuer à élaborer les règles, normes et principes de comportement responsable des États, de discuter des moyens de leur mise en œuvre et d’étudier la possibilité d’établir un dialogue institutionnel régulier avec une large participation sous les auspices des Nations unies ».

Comment le GTCNL était-il organisé ? Le Groupe de travail a été organisé autour d’une session organisationnelle qui a examiné les procédures et le mode de fonctionnement, ainsi que des sessions de fond consacrées à la question, complétées par des réunions intersessions et des assemblées publiques. Le Groupe de travail a tenu dix sessions de fond au cours de son mandat de cinq ans, la onzième et dernière session étant prévue en juillet 2025, lors de laquelle le groupe adoptera son rapport final.

Le GTCNL vu par des experts : réalisations, lacunes et objectifs futurs

Alors que le processus du GTCNL 2019-2025 touche à sa fin, nous avons discuté avec des experts en cybersécurité afin de réfléchir à son impact et d’envisager l’avenir. Leurs réflexions portent sur quatre questions clés :

Les contributions les plus substantielles et les lacunes du GTCNL en matière de sécurité mondiale des TIC
Les priorités pour les futurs dialogues sur le comportement responsable des États dans le cyberespace
La faisabilité d’un consensus sur un mécanisme multilatéral permanent
La pertinence potentielle d’un tel mécanisme dans le climat géopolitique actuel, marqué par des divisions

Leurs points de vue éclairent les résultats obtenus par le GTCNL et les défis qui restent à relever en matière de gouvernance internationale du cyberespace.

Outre des experts externes en cybersécurité, nous avons demandé à notre propre équipe, qui suit le processus du GTCNL depuis sa création, de partager son analyse. Elle met en évidence les principales réalisations de ces cinq dernières années, identifie les lacunes dans les discussions et formule des prévisions sur l’orientation que pourraient prendre les débats lors de la session finale et au-delà.

Lisez l’intégralité des commentaires des experts sur notre page web dédiée.

Souhaitez-vous en savoir plus sur l’IA, les technologies et la diplomatie numérique ? Si oui, interrogez notre chatbot Diplo!

Les cryptomonnaies « mème » : gains rapides ou jeu de hasard?

Jeu numérique à haut risque, les « jetons crypto » brouillent la frontière entre divertissement viral et risque financier dans le domaine le plus sauvage de la cryptomonnaie.

Tape, Face, Head, Person, Animal, Cat, Mammal, Pet

Les cryptomonnaies « mème » ont connu une explosion sur le marché des cryptomonnaies, attirant les investisseurs avec la promesse de profits rapides et un engouement viral. Ces jetons numériques, souvent inspirés de mèmes Internet et de la culture pop, comme Dogecoin, Pepe, Dogwifhat et, plus récemment, Trump coin, n’ont généralement pas d’utilité claire. Leur valeur dépend principalement du buzz sur les réseaux sociaux, du soutien d’influenceurs et de l’enthousiasme de la communauté. En 2025, les cryptomonnaies « mème » restent une tendance controversée mais dominante dans le domaine du trading de cryptomonnaies.

Viral mais vulnérable : l’essor des cryptomonnaies « mème »

Les cryptomonnaies « mème » sont généralement créées à des fins humoristiques, pour susciter l’engagement social ou pour surfer sur les tendances virales sur Internet, plutôt que pour résoudre des problèmes concrets. Malgré cela, elles sont largement connues pour leur popularité et leur attrait massif en ligne. De nombreux investisseurs sont séduits par les cryptomonnaies « mème » en raison de leur potentiel de rendements rapides et importants.

Par exemple, les mème coins sur le thème de Trump ont connu une croissance explosive au début de l’année 2024, le MAGA mème coin (TRUMP) dépassant brièvement les 500 millions de dollars de capitalisation boursière, bien qu’il n’offre aucune utilité réelle et soit largement porté par l’engouement politique et le buzz des réseaux sociaux.

Les rapports d’analyse indiquent qu’en 2024, entre 40 000 et 50 000 nouveaux jetons mème ont été lancés chaque jour, ce chiffre pouvant atteindre 100 000 lors de pics viraux. Solana arrive en tête de la liste des blockchains pour l’activité des mème coins, générant entre 17 000 et 20 000 nouveaux jetons chaque jour.

Le rapport « State of Memecoin 2024 » de Chainplay a révélé que plus de la moitié (55,24 %) des jetons mème analysés étaient classés comme « malveillants ».

Derrière les mème coins : le vrai danger des arnaques

Derrière leur humour et leur attrait viral, les cryptomonnaies « mème » cachent souvent de sérieux risques structurels. Beaucoup sont lancées par des développeurs peu ou pas responsables, et la plupart fonctionnent avec des réserves de liquidités centralisées contrôlées par un petit nombre de portefeuilles. Ce système permet aux créateurs ou aux premiers détenteurs de retirer des liquidités ou de vendre de grandes quantités de jetons sans avertir, ce qui entraîne des effondrements de prix dévastateurs, communément appelés « rug pulls ». (escroquerie au retrait)

Les données sur la chaîne révèlent régulièrement qu’une poignée de portefeuilles contrôlent la grande majorité de l’offre des jetons mème nouvellement lancés, ce qui facilite la manipulation du marché et rend la confiance presque impossible. Ces jetons sont rarement audités, manquent de transparence et n’ont souvent pas de feuille de route claire ni d’utilité à long terme, ce qui expose fortement les investisseurs particuliers.

La combinaison d’une demande alimentée par le battage médiatique et d’une tokenomique opaque fait des jetons mème un terrain favorable à la fraude et à la manipulation, ce qui érode encore davantage la confiance du public dans l’écosystème cryptographique au sens large.

Le jeu déguisé en investissement : la montée d’adrénaline des cryptomonnaies « mème »

Les cryptomonnaies mème exploitent un état d’esprit qui s’apparente davantage au jeu qu’à l’investissement traditionnel. Toute la culture qui les entoure se nourrit d’une spéculation alimentée par l’adrénaline, où chaque hausse de prix donne l’impression de remporter le jackpot et chaque baisse déclenche des montagnes russes émotionnelles à haut risque. Connue sous le nom de « culture degen », (culture assumée de la prise de risque extrême) les traders recherchent des gains rapides, poussés par la peur de passer à côté (FOMO), le battage médiatique et la portée explosive des réseaux sociaux.

Cette mentalité en quête de sensations fortes transforme le trading de mème coins en un jeu de hasard. Les investisseurs prennent souvent des décisions impulsives basées sur le battage médiatique plutôt que sur les fondamentaux, dans l’espoir de profiter d’une hausse soudaine avant le crash inévitable.

Tout est question d’impulsion. Les fluctuations volatiles créent un cycle addictif : l’excitation des gains rapides attire les traders, malgré le risque constant de tout perdre.

Alors que les initiés et les grands détenteurs planifient stratégiquement leurs mouvements pour encaisser le plus possible, la plupart des investisseurs particuliers subissent des pertes, un peu comme des joueurs dans un casino. Le marché des cryptomonnaies mème fonctionne donc moins comme un domaine d’investissement stable que comme un environnement de jeu à haut risque où la chance et le timing l’emportent souvent sur les connaissances et la stratégie.

A chaotic blend of internet culture, greed, and adrenaline, meme coins turn crypto investing into a thrilling game where hype rules and fortunes flip in seconds.

Peut-on vraiment gagner de l’argent avec les mème coins ? Oui, mais à quel prix…

Bien que certains investisseurs aient réalisé des profits substantiels grâce aux cryptomonnaies mème, le succès nécessite des connaissances pointues, des recherches approfondies et une synchronie parfaite. Il est essentiel d’analyser la tokenomics, la croissance de la communauté et les données sur la chaîne avant d’investir. Bien qu’elles puissent être divertissantes, les cryptomonnaies mème constituent un pari risqué. La chance reste un facteur clé, c’est pourquoi les cryptomonnaies mème ne sont jamais considérées comme des investissements sûrs ou à long terme.

Les cryptomonnaies « mème » et le Bitcoin : une histoire de deux états d’esprit

Beaucoup de personnes pensent que toutes les cryptomonnaies partagent la mème philosophie, mais la réalité est tout autre. Il est intéressant de noter que les cryptomonnaies telles que le Bitcoin et les jetons mème reposent sur des philosophies et des motivations psychologiques diamétralement opposées.

Le Bitcoin exprime une philosophie de confiance fondée sur la transparence, la décentralisation et la résilience à long terme. Il séduit ceux qui recherchent la stabilité, la sécurité et une réserve de valeur ancrée dans la technologie et le consensus communautaire : un or numérique qui invite à la patience et à la conviction. En substance, le Bitcoin invite à construire et à conserver avec raison et prévoyance.

Les jetons « mème », par ailleurs, prospèrent grâce à la psychologie de la gratification instantanée, de l’identité sociale et de l’enthousiasme collectif. Ils exploitent notre désir d’excitation, de gains rapides et d’appartenance à un mouvement viral. Leur valeur réside moins dans leur utilité que dans l’émotion partagée : l’espoir, l’engouement et l’adrénaline de surfer sur la prochaine grande vague. Les cryptomonnaies « mème » séduisent par le frisson du moment, le pari et le spectacle social. Elles reflètent ainsi le côté spéculatif et impulsif de la nature humaine, où la frontière entre investissement et jeu est floue.

Comprendre ces fondements psychologiques permet d’expliquer pourquoi les deux coexistent dans le monde des cryptomonnaies, tout en attirant des types d’investisseurs et des mentalités très différents.

Comment les mème coins influencent la réputation de l’ensemble du marché crypto

L’ascension et la chute des cryptomonnaies « mèmes » n’ont pas seulement un impact sur les spéculateurs individuels, elles jettent également une ombre sur la crédibilité de l’ensemble du secteur des cryptomonnaies.

Les escroqueries très médiatisées, les « rug pulls » et les stratagèmes de « pump and dump » (opération de gonflage spéculatif) associés aux jetons « mèmes » érodent la confiance du public et confortent les sceptiques dans leurs inquiétudes. De nombreux courtiers particuliers se lancent dans les cryptomonnaies « mèmes » avec de grands espoirs, mais sont rapidement déçus par les manipulations et les pertes soudaines.

Cela conduit à un sentiment de trahison, qui déclenche une aversion pour le risque et une méfiance généralisée envers tous les actifs cryptographiques, même ceux qui ont des fondamentaux solides comme le Bitcoin ou l’Ethereum. Cette désillusion ne reste pas confinée à une minorité. Elle se propage dans l’opinion publique, décourageant les nouveaux investisseurs et ralentissant l’adoption par les institutions.

À mesure que de plus en plus de personnes associent la cryptomonnaie au jeu et à l’escroquerie plutôt qu’à l’innovation et à la décentralisation, le potentiel de croissance du marché en pâtit. Ainsi, les cryptomonnaies « mèmes », bien qu’elles soient destinées à être des plaisanteries, pourraient avoir de graves conséquences pour l’avenir de la crédibilité de la blockchain.

Des sensations fortes ou une bombe à retardement ?

Les mème coins peuvent offrir des bribes de fortune, mais leur rôle plus occulté dans l’écosystème cryptographique soulève une question qui dérange : sont-ils en train de remodeler la finance ou simplement de la déformer ? Dans un marché où les facéties font bouger des millions et où la spéculation l’emporte sur le fond, le véritable pari n’est peut-être pas seulement financier, mais aussi philosophique.

Sommes-nous en train d’embrasser l’innovation ou de jouer un jeu dangereux avec des dés numériques ? En fin de compte, les mème coins ne sont pas seulement un pari sur les prix, ils sont le reflet du type d’avenir que nous voulons construire dans le domaine des cryptomonnaies. S’agit-il d’une valeur durable ou simplement d’un chaos viral ? La roulette continue de tourner.

Pour plus d’informations sur ces sujets, visitez diplomacy.edu.

Vous souhaitez en savoir plus sur l’IA, les technologies et la diplomatie numérique ? Si oui, interrogez notre chatbot Diplo !

Quand l’IA pense pour nous : quel futur pour notre esprit?

L’essor de l’IA transforme le monde du travail et de l’éducation, mais soulève des questions quant à son impact sur la pensée critique et l’autonomie cognitive.

L’IA réinvente le travail et l’apprentissage

Les progrès rapides de l’IA mettent en lumière toute une série de phénomènes émergents au sein des sociétés humaines contemporaines.

L’intégration d’outils basés sur l’IA dans un large éventail de tâches professionnelles s’est avérée bénéfique à bien des égards, notamment en allégeant la charge cognitive et physique qui pèse traditionnellement sur le travail humain.

En automatisant les processus routiniers et en améliorant les capacités de prise de décision, l’IA a le potentiel d’améliorer considérablement l’efficacité et la productivité dans divers secteurs.

Face à l’accélération des changements technologiques, un nombre croissant de pays accordent la priorité à l’intégration des technologies d’IA dans leurs systèmes éducatifs afin de préparer les élèves aux transformations futures de la société et du monde du travail.

En Chine, l’éducation à l’IA se déploie dès le plus jeune âge

La Chine a publié deux documents politiques historiques visant à intégrer systématiquement l’enseignement de l’IA dans le programme national des écoles primaires et secondaires.

Cette initiative reflète non seulement la vision stratégique à long terme du pays en matière de transformation de l’éducation, mais vise également à positionner la Chine à la pointe de la culture et du développement des talents en matière d’IA à l’échelle mondiale.

chinese flag with the city of shanghai in the background and digital letters ai somewhere over the flag

Ces deux lignes directrices, officiellement intitulées « Lignes directrices pour l’enseignement général de l’IA dans les écoles primaires et secondaires » et « Lignes directrices pour l’utilisation de l’IA générative dans les écoles primaires et secondaires », représentent une approche scientifique et systémique visant à développer les compétences en IA chez les enfants d’âge scolaire.

Leur publication marque une étape importante dans le développement d’un système d’enseignement de l’IA progressif et à plusieurs niveaux, avec des objectifs soigneusement définis en fonction de l’âge et des garanties éthiques pour les élèves et les enseignants.

Les États-Unis développent l’apprentissage de l’IA dans les écoles

En avril, le gouvernement américain a présenté une politique nationale structurée visant à intégrer la maîtrise de l’IA à toutes les étapes du système éducatif.

En créant un groupe de travail fédéral dédié, l’administration entend coordonner les efforts de tous les départements afin de promouvoir un accès précoce et équitable à l’éducation à l’IA.

Au lieu d’isoler l’enseignement de l’IA dans des domaines spécialisés, cette initiative vise à intégrer les concepts de l’IA dans tous les parcours d’apprentissage, de l’enseignement primaire à la formation tout au long de la vie.

Le plan prévoit la création d’un défi national sur l’IA afin d’inspirer l’innovation chez les élèves et les enseignants, en montrant comment l’IA peut répondre à des problèmes concrets.

La politique donne également la priorité à la formation des enseignants afin qu’ils comprennent et utilisent les outils d’IA, au lieu de s’appuyer uniquement sur les méthodes d’enseignement traditionnelles. Elle soutient le développement professionnel afin que les enseignants puissent intégrer l’IA dans leurs cours et réduire la charge administrative.

La stratégie encourage les partenariats public-privé, en s’appuyant sur l’expertise du secteur et les ressources fédérales existantes pour rendre les supports pédagogiques sur l’IA largement accessibles.

L’Union européenne renforce son engagement pour une IA sécurisée

À mesure que l’IA se généralise dans les salles de classe à travers le monde, les enseignants doivent comprendre non seulement comment l’utiliser efficacement, mais aussi comment l’appliquer de manière éthique.

Plutôt que d’introduire des outils d’IA sans accompagnement ni réflexion, la Commission européenne a publié des lignes directrices éthiques pour aider les enseignants à utiliser l’IA et les données de manière responsable dans l’éducation.

Publiées en 2022 et élaborées avec la contribution d’éducateurs et d’experts en IA, les lignes directrices de l’UE s’adressent principalement aux enseignants du primaire et du secondaire qui ont peu ou pas d’expérience en matière d’IA.

Plutôt que de se concentrer sur la complexité technique, elles visent à sensibiliser à la manière dont l’IA peut soutenir l’enseignement et l’apprentissage, à mettre en évidence les risques encourus et à promouvoir une prise de décision éthique.

Les lignes directrices expliquent comment l’IA peut être utilisée dans les écoles, encouragent une utilisation sûre et éclairée par les enseignants et les élèves, et aident les éducateurs à réfléchir aux fondements éthiques de tout outil numérique qu’ils adoptent.

Plutôt que de s’appuyer sur des technologies non éprouvées, elles favorisent une mise en œuvre réfléchie en proposant des questions pratiques et des conseils pour adapter l’IA à divers objectifs éducatifs.

Les outils d’IA pourraient fragiliser la pensée humaine

Cependant, l’augmentation technologique n’est pas sans inconvénients. Des inquiétudes ont été soulevées concernant le risque de perte d’emplois, la dépendance accrue aux systèmes numériques et l’érosion progressive de certaines compétences humaines.

Ainsi, si l’IA offre des opportunités prometteuses pour améliorer le lieu de travail moderne, elle introduit également des défis complexes qui doivent être examinés de manière critique et abordés de manière responsable.

L’un des défis majeurs à relever dans le contexte d’une dépendance croissante à l’IA est le phénomène connu sous le nom de « déchargement cognitif ». Mais que signifie exactement ce terme ?

Et si penser devenait optionnel ?

Le « déchargement cognitif » désigne la pratique consistant à utiliser des actions physiques ou des outils externes pour modifier les exigences d’une tâche en matière de traitement de l’information, dans le but de réduire la charge cognitive d’un individu.

Il s’agit essentiellement de transférer certaines fonctions mentales, telles que la mémoire, le calcul ou la prise de décision, vers des ressources externes telles que des appareils numériques, des notes écrites ou des cadres structurés.

Si cette stratégie peut améliorer l’efficacité et les performances, elle soulève également des questions concernant le développement cognitif à long terme, la dépendance à l’égard des aides technologiques et la dégradation potentielle des capacités mentales innées.

Comment l’IA pourrait affaiblir la pensée critique

Une étude menée par le Dr Michael Gerlich, directeur du Centre pour la prospective stratégique et la durabilité des entreprises à la SBS Swiss Business School, publiée dans la revue Societies, soulève de graves préoccupations quant aux conséquences cognitives de l’augmentation de l’IA dans divers aspects de la vie.

Elle suggère que l’utilisation fréquente d’outils d’IA pourrait affaiblir la capacité des individus à exercer leur esprit critique, une compétence considérée comme fondamentale pour le raisonnement indépendant, la résolution de problèmes et la prise de décisions éclairées.

Plus précisément, le Dr Gerlich a adopté une approche mixte, combinant des données d’enquête quantitative auprès de 666 participants et des entretiens qualitatifs avec 50 personnes.

Les participants, issus de différents groupes d’âge et de différents niveaux d’éducation, ont été évalués sur la fréquence de leur utilisation des outils d’IA, leur tendance à se décharger de tâches cognitives et leurs performances en matière de pensée critique.

L’étude a utilisé des mesures de la pensée critique basées sur l’auto-évaluation et sur les performances, ainsi que des analyses statistiques et des modèles d’apprentissage automatique, tels que la régression aléatoire forestière, afin d’identifier les facteurs clés influençant les performances cognitives.

Plus ils utilisent l’IA, moins les jeunes pensent par eux-mêmes

Les résultats ont révélé une forte corrélation négative entre l’utilisation fréquente de l’IA et les capacités de réflexion critique. Les personnes qui ont déclaré dépendre fortement des outils d’IA, que ce soit pour obtenir des réponses rapides, des explications résumées ou des recommandations algorithmiques, ont obtenu des scores plus faibles aux évaluations de la réflexion critique.

Cet effet était particulièrement prononcé chez les jeunes utilisateurs âgés de 17 à 25 ans, qui ont déclaré les niveaux les plus élevés de déchargement cognitif et ont obtenu les résultats les plus faibles dans les tâches de pensée critique.

En revanche, les participants plus âgés (âgés de 46 ans et plus) ont démontré des capacités de pensée critique plus solides et étaient moins enclins à déléguer leur effort mental à l’IA.

L’enseignement supérieur renforce la pensée critique

Les données ont également montré que le niveau d’éducation constituait un facteur protecteur : les personnes ayant un niveau d’éducation plus élevé faisaient systématiquement preuve de capacités de réflexion critique plus solides, quel que soit leur niveau d’utilisation de l’IA.

Ces résultats suggèrent que l’éducation académique peut doter les individus de meilleurs outils pour aborder de manière critique les informations numériques plutôt que d’accepter sans réserve les réponses générées par l’IA.

Il convient toutefois de noter que, bien que l’étude n’établisse pas de lien de causalité direct, la force des corrélations et la cohérence des données quantitatives et qualitatives suggèrent que l’utilisation de l’IA pourrait effectivement contribuer à un déclin progressif de l’indépendance cognitive.

Cependant, dans son étude, M. Gerlich souligne également la possibilité d’une causalité inverse : les personnes ayant des capacités de réflexion critique plus faibles pourraient être plus enclines à se fier aux outils d’IA.

Moins on pense, moins on retient

Si le déchargement cognitif peut améliorer les performances immédiates, il se fait souvent au détriment de la mémoire à long terme, comme le montrent d’autres études.

Ce compromis s’est illustré de façon particulièrement marquante dans des tâches expérimentales telles que la tâche de copie de motifs, où les participants chargés de reproduire un modèle choisissent généralement de s’y référer à plusieurs reprises plutôt que de le mémoriser.

Même lorsque ces comportements entraînent un surcroît de temps ou d’efforts (par exemple, se déplacer physiquement d’un poste à l’autre), la majorité des participants choisissent de décharger, ce qui suggère une forte préférence pour la minimisation de l’effort cognitif.

Ces résultats soulignent la tendance humaine à privilégier l’efficacité plutôt que l’internalisation, en particulier dans des conditions exigeantes sur le plan cognitif.

La tendance à décharger soulève des questions cruciales sur les conséquences cognitives et éducatives d’un recours prolongé à des aides externes. D’une part, le transfert peut libérer des ressources mentales, permettant aux individus de se concentrer sur la résolution de problèmes plus complexes ou sur plusieurs tâches à la fois.

D’autre part, il peut favoriser une sorte de dépendance cognitive, affaiblissant les traces mémorielles internes et réduisant les possibilités d’engagement profond avec l’information.

Dans ce cadre, le transfert cognitif n’est pas un échec de la mémoire ou de l’attention, mais une reconfiguration de l’architecture cognitive, un processus qui peut être adaptatif plutôt que préjudiciable.

Toutefois, ce point de vue reste controversé, notamment à la lumière des conclusions de l’étude de Gerlich, qui montre que le déchargement fréquent peut nuire à la mémorisation, au transfert des connaissances et à la pensée critique.

Si les étudiants, par exemple, s’appuient continuellement sur des appareils numériques pour se souvenir de faits ou résoudre des problèmes, ils risquent de ne pas développer les modèles mentaux solides nécessaires à un raisonnement flexible et à une compréhension conceptuelle.

L’esprit est-il plus vaste que le cerveau ?

Cette tension a également suscité un vif débat parmi les scientifiques cognitifs et les philosophes, notamment à la lumière de l’hypothèse de l’esprit étendu. Contrairement à la vision traditionnelle qui limite la cognition au cerveau, cette théorie soutient que les processus cognitifs s’appuient souvent sur des outils, des environnements et des structures sociales, auxquels ils sont étroitement liés et distribués.

À mesure que les technologies numériques s’amalgament de plus en plus à notre quotidien, cette hypothèse soulève des questions fondamentales sur l’identité humaine, la cognition et l’action.

Au cœur de la thèse de l’esprit étendu se trouve une question d’une simplicité trompeuse : où s’arrête l’esprit et où commence le reste du monde ?

En établissant une analogie avec les prothèses, des objets externes qui deviennent fonctionnellement partie intégrante du corps, Clark et Chalmers soutiennent que les outils cognitifs tels que les ordinateurs portables, les smartphones et les carnets de croquis peuvent devenir des composants intégrés de notre système mental.

Ces outils ne se contentent pas de soutenir la cognition ; ils la constituent lorsqu’ils sont utilisés de manière transparente et fonctionnellement intégrée. Ce changement conceptuel a redéfini la pensée non plus comme un processus limité au cerveau, mais comme une interaction dynamique entre l’esprit, le corps et le monde.

Entre humain et machine : l’enjeu de l’équilibre

En conclusion, le déchargement cognitif représente un mécanisme puissant de la cognition moderne, qui permet aux individus de s’adapter à des environnements complexes en répartissant la charge mentale.

Cependant, ses effets à long terme sur la mémoire, l’apprentissage et la résolution de problèmes font encore l’objet de recherches actives. Plutôt que de considérer le déchargement comme intrinsèquement bénéfique ou nuisible, les recherches et les pratiques futures devraient chercher à équilibrer son utilisation, en tirant parti de ses avantages tout en atténuant ses inconvénients.

Human VS Ai Background Brain and heart hd background 1024x576 1

En fin de compte, en tant qu’ éducateurs, décideurs politiques et technologues, nous devons façonner l’avenir de l’apprentissage et du travail, tout en faisant face à une tension fondamentale : comment tirer parti des avantages de l’IA sans compromettre les facultés mêmes qui définissent l’intelligence humaine, à savoir la pensée critique, la mémoire et le jugement indépendant.

Pour plus d’informations sur ces sujets, visitez diplomacy.edu.

Souhaitez-vous en savoir plus sur l’IA, les technologies et la diplomatie numérique ? Si oui, interrogez notre chatbot Diplo !

DW Weekly #220 – Diplo reporting from WSIS+20 High-Level Event in 2025, UN OEWG 2021-2025 11th substantive session reports, and more digital policy news!

4 – 11 July 2025

People, Person, Book, Publication, Adult, Male, Man, Advertisement, Poster, Baby, Comics, Head, Face, Art

Dear readers,

This week, cybersecurity faced heightened challenges as global instability, including conflicts like Israel-Palestine and US-Iran tensions, fueled a surge in cyberattacks, as per the report by GlobalData, which shows that rising geopolitical tensions are giving state actors, terrorists, hacktivists and cybercriminals more opportunities to strike, targeting critical infrastructure and escalating risks for businesses worldwide.

Brazil, for example, suffered its largest bank breach, with hackers stealing $180 million, while South Korea’s SK Telecom announced a $700 billion upgrade to bolster its cybersecurity defences.

The EU regional authorities called for greater involvement in EU cybersecurity plans, emphasising the need for local expertise to shape effective hospital protection strategies. Another alert: the FBI recently warned of rising airline cyberattacks posing national security threats.

Ukraine is strengthening cybersecurity ties with the EU amid ongoing threats. Discussions in Rome focused on enhancing cybersecurity cooperation and establishing a joint Center of Competence for Cyber Resilience in Ukraine.

Visa’s war room leverages AI to combat $15 trillion in threats, and the US Cyber Command proposed a $5 million AI initiative for 2026, a project aimed at developing data standards and pilot technologies that enhance threat detection, automation, and decision-making in cyber operations.

In the lead-up to the concluding session of the UN Open-Ended Working Group (OEWG) on ICTs, Thailand, New Zealand, and South Korea have released their respective national positions on the application of international law in cyberspace, contributing to the growing corpus of state practice on the issue.

AI in education

A university lecturer in the US says many students increasingly rely on ChatGPT to write essays, even about the ethics of AI, raising concerns about critical thinking in higher education. On the other hand, ChatGPT has ‘quietly’ rolled out the feature ‘Study Toge ther.’ Users have noticed the new option ‘Study Together’ appearing among available tools, though OpenAI has yet to confirm any official rollout. The feature seems designed to make ChatGPT a more interactive educational companion than just delivering instant answers.

Crypto developments

A new analysis by crypto commentator Stellar Rippler suggests that Donald Trump’s Big Beautiful Bill may be part of a calculated effort to dismantle the current financial order. Meanwhile, Bitcoin is holding firm above $108,000, trading at $108,387 as of 8 July. Despite a slight daily dip, longer-term indicators support a bullish trend. On the other side of the Atlantic, more than 50 crypto firms are now fully licensed under the EU’s MiCA framework, six months after it came into effect. The list names 14 stablecoin issuers and 39 service providers approved to operate across the EU’s 30 member states, except Tether and Binance, which remain absent from the approved list.

Chip export control side effects

The US is preparing stricter export controls on high-end Nvidia AI chips destined for Malaysia and Thailand, in a move to block China’s indirect access to advanced GPU hardware. At the same time, Samsung Electronics expects its second-quarter operating profits to exceed half, citing Washington’s export controls on advanced AI chips to China. The company announced a projected 56% year-on-year drop in operating profit, falling to 4.6 trillion won ($3.3 billion), with revenue down 6.5% from the previous quarter.

Data and content governance

Data and content governance saw significant developments this week as BRICS nations pushed for AI and data regulations amid de-dollarisation challenges, highlighting global governance tensions. Still speaking about data and content, a court ruling raised alarms over saved ChatGPT chats, exposing privacy risks, while Elon Musk ’s Grok AI chatbot removed offensive content to align with ethical standards, though it faced suspension in Türkiye following a court order.

Diplo Blog – Diplomacy is where there are no rules

Aldo Matteucci

He had bought a large map representing the sea,
Without the least vestige of land:
And the crew were much pleased when they found it to be
A map they could all understand.

–Lewis Carroll, The Hunting of the Snark

I’d made this comment recently in an off-hand sort of way. I’ve been challenged, so I must stand by my words: diplomacy is where there are no rules.

Let me explain the assertion with an analogy. At the dawn of the electronic age, gadgets were all stand-alone. They were not able to interconnect and communicate. Frustration all around. Slowly, protocols were established, growing into full-fledged interfaces. Now, it’s: ‘plug and play’. Until an agreement was reached on how to connect the two pieces of equipment, there were no rules – just an unexplored set of possibilities. Read the full blog!

Last week in Geneva

WSIS+20 High-Level Event in 2025

The annual WSIS Forum, branded as the WSIS+20 High-Level Event in 2025, kicked off on 7 July in Geneva. The event served as a platform for multistakeholder discussions on implementing WSIS Action Lines and related progress, challenges, and opportunities since adopting the Geneva Plan of Action in 2003. In the lead-up to the WSIS+20 review by the UN General Assembly, the event also featured discussions on progress made in implementing the WSIS outcomes.

Helmet, People, Person, Adult, Male, Man, Face, Head, American Football, Football, Playing American Football, Sport, Crowd, Clothing, Formal Wear, Suit, Furniture, Table, Bill Kenney

Diplo has actively participated again, reporting just-in-time and spreading the diplomatic word as ever, with Dr Jovan Kurbalija, Head of Diplo and Geneva Internet Platform, who focused on how digital tools are reshaping diplomacy by fostering inclusivity and efficiency, urging a shift from reactive to proactive governance. He advocated mapping digital policy complexities—over 1,000 mechanisms tracked by the Geneva Internet Platform—to avoid duplication and enhance collaboration, proposing policy incubators and cooperation accelerators to connect diverse stakeholders. His vision, rooted in human-centric approaches, calls for balancing innovation with ethical oversight, a theme resonating throughout the event.

The event’s focus on inclusion, language, ethics, cybersecurity, sk i lls, and humanitarian efforts reflects a holistic agenda, but also critical gaps such as funding disparities, regulatory weaknesses, and geopolitical divides, challenging the narrative of a unified digital future. The GDC synergy offers potential, yet its success hinges on addressing these structural issues and ensuring digital diplomacy evolves into actionable, equitable progress rather than remaining a well-intentioned dialogue.

AI for Good Global Summit 2025

Crowd, Person, Audience, Electrical Device, Microphone, Boy, Child, Male, Accessories, Glasses, People, Speech, Computer, Electronics, Laptop, Pc, Clothing, Formal Wear, Suit, Head, Podium, Face, Carlos Julio Arosemena Monroy

AI for Good Global Summit 2025, the flagship event of the ITU-led AI for Good platform, organised in partnership with 40 UN agencies and co-convened with Switzerland, kicked off on 8 July in Geneva. This year’s programme featured three key events: the AI for Good Global Summit (8–9 July), AI Governance Day (10 July), and International AI Standards Day (11 July).

Both events wrap up later today. From hosting sessions and exhibiting to sharing our expertise on stage, our Diplo and GIP team was active throughout both events. Visit our dedicated WSIS+20 High-Level Event 2025 and AI for Good Global Summit 2025 web pages on diplomacy.edu for more information about our activities.

Plus, don’t miss our session reports and AI insights—available now on our dedicated WSIS+20 High-Level Event 2025 and AI for Good Global Summit 2025 web pages on the Digital Watch Observatory.

In Case You Missed It
The OEWG on ICT security is concluding its work this week with its eleventh and final substantive session. We’ve been following developments closely with session reorts and will publish a full analysis next week—stay tuned.

Upcoming events

For the main updates, reflections and events, consult the RADAR, the READING CORNER and the UPCOMING EVENTS section below.

Join us as we connect the dots, from daily updates to main weekly developments, to bring you a clear, engaging monthly snapshot of worldwide digital trends.

DW Team

RADAR

Highlights from the week of 4 – 11 July 2025

Parliamentarians step up as key players in shaping the digital future

From regulating AI to bridging the global digital divide, parliamentarians are redefining their roles to ensure that technology serves people, not the other way around.

Report shows China outpacing the US and EU in AI research

Governments now face the reality that falling behind in AI capability could have serious geopolitical consequences, warns a new research report.

technology network background connection cyber space ai generative

Space operators face strict cybersecurity obligations under EU plan

EU’s law aims to protect supply chains and ground systems supporting space missions.

Harnessing the power of space: Bridging innovation and the SDGs

From legal loopholes to breakthrough technologies, the global space community is charting a bold new course to make outer space a powerful ally in solving Earth’s most urgent challenges.

Microsoft Salesforce AI call centres layoffs jobs

Microsoft and Salesforce use AI to cut costs and reshape workforce

Microsoft and Salesforce report major cost savings and operational changes as AI replaces a significant portion of human-driven tasks.

EU rejects delay for AI Act rollout

Tech giants asked for a delay, but the EU is sticking to its original AI Act deadlines through 2026.

Kurbalija: Digital tools are reshaping diplomacy

Geneva’s timeless role in global affairs is now expanding into the digital realm, where the future of diplomacy hinges on how we govern the technologies reshaping our world.

Google hit with EU complaint over AI Overviews

EU publishers accuse Google of misusing content in AI Overviews, claiming serious damage to their online visibility and income.

meta brazil hate speech policy 1024x673 1

Meta hires Apple’s top AI executive amid tech talent war

The departure of Apple’s foundation models leader reflects internal instability and Meta’s growing influence in AI development.

Wimbledon faces backlash over AI line judges after tech errors spark outrage

Jack Draper and Emma Raducanu criticised AI calls at Wimbledon, saying they led to key point losses and unfair match results.

xAI unveils Grok 4 with top benchmark scores

Elon Musk revealed Grok 4, claiming it surpasses PhD-level knowledge, despite recent backlash over offensive AI-generated content.

quantum ai data science and cybersecurity

Quantum leap in Finland pushes qubit to millisecond range

Aalto and VTT researchers report over 1 millisecond coherence, a breakthrough for superconducting quantum devices.

Google partners with UK government on AI training

Collaboration with Google DeepMind will support departments in deploying AI solutions across the public sector.

AI that serves communities, not the other way round

Shifting the focus from powerful algorithms to local voices, a global panel explored how AI can thrive in the Global South without leaving its communities behind.

View all updates

READING CORNER

The power of self-organisation

Can Wikipedia teach diplomacy a lesson? Aldo Matteucci contrasts rigid hierarchies with messy, adaptive self-organising systems, and asks which one truly gets more done.

AI Apprenticeship for IOs · From diplomats to AI builders

The AI Apprenticeship for International Organisations, developed by DiploFoundation, empowers professionals from entities like the UN, WHO, and CERN to create AI tools that enhance global cooperation.

Diplomacy is where there are no rules

Diplomacy begins where there are no rules – only open-ended possibilities. Aldo Matteucci explains how real diplomacy begins: with trial-and-error, creative thinking, and turning confusion into agreement.

BLOG featured image 2025 Why military AI needs urgent regulation

Why military AI needs urgent regulation – Diplo

As military AI becomes operational in today’s conflicts, the lack of regulation and accountability risks turning warfare into a domain governed by opaque algorithms and unchecked escalation.

The rise and risks of synthetic media

Balancing synthetic media’s benefits with its risks will shape the future of digital communication and societies.

16 July 2025

CPDP LatAm session on ‘Critical infrastructure and digital sovereignty: regulatory challenges for sustainable data centers’

The growing global demand for the Internet and digital services has consolidated data centers as critical infrastructures for international trade, technological innovation, environmental sustainability and digital sovereignty, highlighting their strategic centrality in the digital economy.

14 Jul 2025 – 23 Jul 2025

2025 High-Level Political Forum on Sustainable Development (HLPF)

The 2025 High-Level Political Forum on Sustainable Development (HLPF), under the auspices of the United Nations Economic and Social Council (ECOSOC), will occur from 14 July to 23 July 2025.

June 2025 in Retrospect

Dear readers,

In June, the spotlight shone on the 20th Internet Governance Forum (IGF) 2025 in Lillestrøm, Norway, where DiploAI made waves as the official reporting partner, capturing every session with cutting-edge AI-driven transcription and delivering real-time insights via dig.watch.

Amid global debates on AI governance, digital sovereignty, and cybersecurity, the forum set the stage for transformative dialogue. From bridging the digital divide to tackling AI-driven disinformation, June’s trends reveal a world at a crossroads, where technology shapes geopolitics, trade, and human rights.

Join us as we unpack the key developments of the month, spotlighting the pivotal moments of IGF 2025 and the broader digital diplomacy landscape.

Some digital highlights of June 2025:

IGF 2025 opens in Norway: the forum kicked off on 23 June, bringing together over 8,000 stakeholders to shape internet governance.
EU allocates €145.5 million for cybersecurity – funds target healthcare and critical infrastructure protection.
Bitcoin holds above $100,000 – crypto stability signals growing adoption amid regulatory shifts.
NATO summit addresses cyber threats – pro-Russian hacktivists target infrastructure during 76th summit.
HRC reviews tech and human rights – reports on AI and digital birth registration debated on 26 June.
ITU Council adopts 2026–2027 budget and focuses on digital inclusion and AI diplomacy in Geneva.
WhatsApp is banned on US House of Representatives devices – Meta’s service is restricted due to security concerns.
Quantum computer orbits via SpaceX – a photonic breakthrough hints at future tech shifts.
The EU urges a delay of the AI Act, as the industry cites missing frameworks and legal uncertainty.
Global South pushes for digital inclusion – IGF session highlights governance for the vulnerable.

Diplo’s analysis and reporting in an exceptional time

In a world where history unfolds at breakneck speed, the real challenge isn’t just keeping up—it’s making sense of it all. Every day brings a flood of information, but the bigger picture often gets lost in the noise. How do today’s developments shape long-term trends? How do they impact us as individuals, communities, businesses, and even humanity?

At Diplo, we bridge the gap between real-time updates and deeper insights. Our Digital Watch keeps a pulse on daily developments while connecting them to weekly, monthly, and yearly trends as illustrated bellow.

From cybersecurity to e-commerce to digital governance, we track these shifts from daily fluctuations to long-term industry pivots.

In our 101st issue of the monthly newsletter, you can follow: AI and tech TENDENCIES | Developments in GENEVA | dig.watch ANALYSIS

Best regards,

DW Team

June in Retrospect: IGF 2025 and digital trends

June 2025 was defined by the resounding impact of the 20th Internet Governance Forum (IGF) 2025, held on 23 – 27 June in Lillestrøm, Norway, at Nova Spektrum, the country’s largest UN gathering ever hosted. With over 4,000 participants joining in person and online, the forum, themed ‘Building Digital Governance Together’, was opened by UN IGF Secretariat Head Chengetai Masango and Norway’s Minister of Digitalisation Karianne Oldernes Tung, stressing a collaborative vision for an open, secure internet amid rising global tensions. Key sessions on Day 1 tackled AI’s role in humanity and the digital divide, with 2.6 billion still offline, while Dr Jovan Kurbalija called for a renewed IGF mandate and innovative funding ahead of WSIS+20, setting a foundation for inclusive governance.

The forum’s dialogue deepened on Day 2 with showcases of India’s Aadhaar (80 million daily uses) and Brazil’s PIX ($5.7 billion savings). However, funding challenges were stark, with Dr Kurbalija and Sorina Teleanu advocating for AI cooperation and parliamentary engagement. Days 3 and 4 focused on cultural diversity in AI governance and the knowledge ecology project, unlocking 19 years of IGF data for SDGs, while the final day addressed internet fragmentation under Article 29C of the Global Digital Compact (GDC), with Marilia Maciel urging economic research and Gbenga Sesan highlighting user disparities. Key takeaways included a push for measurable frameworks, sustained multistakeholder collaboration, and DiploAI’s pivotal role in delivering real-time, AI-driven reporting via dig.watch, underscoring June as a crucial month for digital policy.

Rising AI governance frameworks

June witnessed a decisive surge in AI governance frameworks, crystallised through converging institutional, regional, and civil-society efforts showcased at dig.watch. At IGF 2025 in Lillestrøm, sessions like the High-Level Review of AI Governance identified a troubling concentration of compute in a small group of nations, calling for democratised GPU access, algorithmic audits, and robust model evaluation to combat 26 % hallucination rates and biassed outcomes. In parallel, the UNESCO Global Forum on the Ethics of AI reaffirmed its commitment to the 2021 Recommendation, bringing into focus human rights, gender sensitivity, multistakeholder alignment, and sustainability outcomes. IGF workshops like AI Innovation Responsible Development Ethical Imperatives stressed global ethical unity and emphasised inclusive public systems, from agriculture to education, to ensure AI serves broad social goals.

Meanwhile, IGF’s Open Forum #33, facilitated by China’s Cyberspace Administration, echoed the need for equitable cooperation, spotlighting that over 60 % of AI patents are Chinese and underscoring the importance of shared regulatory norms and public interest safeguards. Sessions such as AI at a Crossroads Between Sovereignty and Sustainability further fused debates on environmental justice and strategic autonomy, examining the unsustainable water use of data centres and the resource-linked vulnerabilities of national AI infrastructures.

Beyond global summits, dig.watch also flagged UNGA’s revision of an AI Governance Dialogue draft on 4 June as a pivotal move toward inclusive science diplomacy—the revised text mandates balanced leadership and annual assembly-level reporting. AI capacity-building platforms like UNESCO’s 4–5 June Paris forum and ITU Council workshops underscore the need for investments in governance training and readiness for public-sector AI deployment.

Sovereignty in the digital age – data governance and justice imperatives

In Europe’s heart, Denmark’s Ministry of Digitalisation initiated a government-wide migration from Microsoft products toward LibreOffice and Linux, echoing parallel efforts in Copenhagen and Aarhus to strengthen national autonomy and data control over US-based platforms. Not far behind, the German state of Schleswig‑Holstein formally abandoned Microsoft Teams and Office, citing geopolitical vulnerabilities and licensing inequities, while redirecting data storage to German-run infrastructures. These synchronised moves come as Lyon, the third-largest city in France, is also transitioning its municipal systems to open-source stacks like Nextcloud and PostgreSQL, exemplifying a broader municipal sovereignty wave that signals Europe’s collective pivot in public sector IT.

Concurrently, the EU launched DNS4EU on 9 June, establishing a privacy-centric, GDPR-compliant Domain Name System resolver managed by a pan-European consortium to wean the region off US-dominated DNS services and fortify continental infrastructure. In the United Kingdom, tech leaders are reevaluating reliance on American cloud providers, with most IT professionals supporting ‘cloud repatriation’ to reclaim data sovereignty and safeguard strategic autonomy under the CLOUD Act shadow.

At IGF 2025 in Lillestrøm, community networks and open-source platforms emerged as central to sovereignty discussions, highlighted at four daily sessions that elevated grassroots-managed infrastructure to balance centralised, corporate-controlled systems. This localist strand was reflected globally in sessions like AI at a Crossroads Between Sovereignty and Sustainability, where speakers exposed the ecological interdependencies—such as water-intensive data centres—and advocated for a ‘digital solidarity’ model prioritising shared Global South priorities over resource-driven fragmentation.

Data governance is intertwined with justice, as IGF 2025’s Day 3 emphasised cultural diversity in AI, with the knowledge ecology project aiding SDGs. ‘Civil society pushes for digital rights and justice in WSIS+20 review’ demanded accountability, and the HRC’s tech report (A/HRC/59/32) addressed AI’s human rights impact.

Geopolitical cyber tensions

June brought major escalations in geopolitical cyber tensions, with events converging to reveal a global landscape where digital infrastructure now sits at the heart of strategic conflict.

Japan’s introduction of the Cyber‑Defense Act early in the month symbolised a paradigm shift: national cyber sovereignty is now defence policy, empowering pre-emptive actions against hostile servers before they can strike critical infrastructure. Across the Atlantic, NATO’s summit took a transformative tone; member states not only agreed to expand defence budgets to 5% of GDP but formally acknowledged cyberespionage and sabotage by Russia-linked groups as frontline threats, prompting joint exercises to reinforce cyber-resilience across allied networks.

At the same time, a surge in hacktivist operations tied to Iran–Israel tensions signalled a volatile expansion of cyber conflict zones. A dig.watch cybercrime summary noted a sharp uptick in attacks on airlines and government systems, particularly amid the Israel-Iran crisis.

Beyond state-led conflict, cyber threats intensified at the societal level, from airports to healthcare facilities and new malware campaigns exploiting ubiquitous platforms like Zoom to compromise crypto assets.

Crypto’s regulatory evolution

June underscored a rapid evolution in crypto regulation and adoption: shifts in concrete regulatory action, market shifts, and mainstream acceptance. In Europe, the long-anticipated MiCA framework edged closer to implementation by 2025, signalling the EU’s intent to regulate stablecoins, enhance transparency, and protect consumers. In Korea, intense political and legislative momentum accompanied a new bill empowering commercial entities to issue stablecoins.

Across the Atlantic, the USA experienced dramatic developments: Bitcoin surged past USD 100,000, buoyed by policy signals and broader institutional interest. Simultaneously, plans by the Federal Housing Finance Agency to permit crypto holdings in mortgage reserve assets—nurtured by presidential momentum—highlighted growing regulatory curiosity from the heart of the financial system.

Regulators also ramped up consumer protections and enforcement: Barclays in the UK blocked crypto transactions on credit cards over consumer risk concerns, while Singapore’s MAS demanded local crypto firms cease overseas digital token services by 30 June, or face fines and jail, strengthening KYC and AML protocols.

Overlaying these trends, quantum computing’s impact on crypto security loomed large: experts warned that advances like Microsoft’s Majorana chip could crack Bitcoin within five years, highlighting a looming existential challenge to current blockchains and urging urgent investment in quantum-resistant systems.

Trade and AI chip control battles – export controls as cybersecurity and privacy fortification

Taiwan imposed urgent export controls on TSMC, stemming from its inadvertent sale of chiplets to Huawei—an incident that compelled Taipei to heighten scrutiny across advanced semiconductor transfers. Simultaneously, the UAE’s Stargate AI megaproject, facing US export concerns due to proximity to sanctioned regimes, illustrated how non-Western AI investments are now contingent on compliance with geopolitically sensitive terms.

Bridging the digital divide and digital rights

The IGF 2025 in Lillestrøm started with sobering statistics: 2.6 billion people remain offline, an alarming figure reframing connectivity as a democratic imperative, not just a technical goal. Sessions like Closing Digital Divides by Universal Access & Acceptance brought together voices from Canada, Kenya, and Pakistan to explore socioeconomic and linguistic barriers blocking inclusion (turn0search0). The UNESCO-endorsed ROAMX framework was spotlighted for its role in evaluating digital policy through a lens of Rights, Openness, Accessibility, Multistakeholder participation, and gender and sustainability considerations.

Digital rights gained focus, with IGF 2025’s civil society push for WSIS+20 accountability and HRC’s emphasis on marginalised groups. ‘Civil society pushes back against cyber law misuse’ highlighted government overreach, while the Freedom Online Coalition’s AI statement upheld rights, although spyware gaps in the Global South persisted.

Language diversity also took centre stage, with an IGF panel emphasising that true connectivity requires a native-language interface, a plea backed by experts from ICANN, Unicode, and regulatory authorities across regions.

On the policy front, the June release of the WSIS+20 Elements Paper emphasised equitable access, data justice, cybersecurity, and inclusive AI as pillars of global digital cooperation.

Geneva’s own Canton is implementing a constitutionally enshrined right to digital integrity, diving into operational standards for data protection and anti-surveillance—an experimental testbed in Europe’s digital rights frontier (turn0search17). Meanwhile, an IGF roundtable on ‘Click with Care’ recombined child safety, algorithmic impact, and hate speech concerns into a unified digital rights narrative, provoking parliamentarians’ calls for platform accountability laws and human-centred design.

June 2025’s trends, from IGF 2025’s insights to global shifts, shape our digital future. Explore more on dig.watch!

Diplo Blog – ‘AI and Magical Realism: When technology blurs the line between wonder and reality’

The challenges of governing AI often feel like something out of a Gabriel García Márquez novel, where the extraordinary blends seamlessly with the everyday, and the line between the possible and the impossible grows faint. In the week from 22 to 27 June 2025 at the 20th Internet Governance Forum (IGF), I proposed using magical realism as a lens to understand AI’s complexities. Here’s why this literary tradition might offer a useful tool for the AI debates ahead.

Read the full blog on diplomacy.edu!

Join us next month as we track these evolving trends. Subscribe to our weekly updates at dig.watch for the latest digital policy insights!

For more information on cybersecurity, digital policies, AI governance and other related topics, visit diplomacy.edu.

Geneva in June

Developments, events and takeaways

In June, Geneva reaffirmed its position as a global nerve centre for digital policy, offering a rich tapestry of events, summits, and multilateral initiatives that articulated the future of digital governance.

The month began with the Giga School Connectivity Forum on 5 June at Campus Biotech, which brought together cross-sector stakeholders, including government, academia, and telecommunications leaders, to dissect challenges in achieving universal school internet access by 2030, emphasising the urgency of closing educational and digital gaps. Shortly thereafter, the Final Brief on WSIS+20 High-Level Event on 10 June, convening diplomats, UN officials, and technologists in Geneva to outline priorities for the December 2025 UN General Assembly review, on issues such as digital inclusion, cybersecurity, and inclusive AI frameworks, cementing the city’s role in shaping global digital norms.

Mid-June saw ITU Council session (17–27 June) reconvene at ITU headquarters. The ‘AI in Action’ workshop and discussions on the Giga Connectivity Centre, launched in partnership with UNICEF and Spain at Campus Biotech, highlighted how AI and connectivity are being mainstreamed into Geneva’s institutional agenda. Simultaneously, on 9 June, the DNS4EU initiative captured attention, even though it is a separate project, its launch underlines Geneva’s alignment with European infrastructure sovereignty goals.

Geneva is also hosting the 59th session of the UN Human Rights Council, running from 16 June to 11 July. Alongside traditional rights issues, delegates tackled digital-era concerns: the impact of AI on freedom of expression, surveillance, and the rights of vulnerable populations. While digital-specific resolutions were not yet tabled, the session reaffirmed that the Council considers digital rights within its broader human rights framework, setting the stage for deeper engagement on online privacy, misinformation, and digital inclusion in upcoming global reviews.

From 19 to 30 June, the Geneva Internet Law Summer School welcomed global scholars to debate issues ranging from data protection to net neutrality, underscoring Geneva’s status as a laboratory for digital legal policy. In parallel, Geneva-based ARTICLE 19 led a powerful session demanding digital rights and equity to be central to the WSIS+20 outcome, framing digital inclusion as an issue inseparable from human rights.

Snapshot: The developments that made waves in June

June 2025 was marked by notable developments in AI governance, cybersecurity, and global digital policy. Here’s a snapshot of what happened over the last month:

TECHNOLOGY

Chinese chipmaker Loongson has unveiled new server CPUs that it claims are comparable to Intel’s 2021 Ice Lake processors, marking a step forward in the nation’s push for tech self-sufficiency.

Africa is falling far behind in the global race to develop AI, according to a new report by Oxford University.

A groundbreaking quantum leap has taken place in space exploration. The world’s first photonic quantum computer has successfully entered orbit aboard SpaceX’s Transporter 14 mission.

The US Department of Defence has awarded OpenAI a $200 million contract to develop prototype generative AI tools for military use.

Recent breakthroughs in quantum computing have revived fears about the long-term security of Bitcoin (BTC).

A small-scale quantum device developed by researchers at the University of Vienna has outperformed advanced classical machine learning algorithms—including some used in today’s leading AI systems—using just two photons and a glass chip.

Orange Business and Toshiba Europe have launched France’s first commercial quantum-safe network service in Paris.

Oxford University physicists have achieved a world-first in quantum computing by setting a new record for single-qubit operation accuracy.

Oxford Quantum Circuits (OQC) has revealed plans to develop a 50,000-qubit fault-tolerant quantum computer by 2034, using its proprietary ‘Dimon’ superconducting transmon technology.

Chinese scientists have created the world’s first AI-based system capable of identifying real nuclear warheads from decoys, marking a significant step in arms control verification.

GOVERNANCE

The new EU International Digital Strategy 2025 (published on 5 June 2025) pivots from the EU’s values-based digital diplomacy towards a more geopolitical, security – and competition–driven approach.

The US Senate has passed the GENIUS Act, the first bill to establish a federal framework for regulating dollar-backed stablecoins.

Vietnam has officially legalised crypto assets as part of a landmark digital technology law passed by the National Assembly on 14 June.

INFRASTRUCTURE

The UK government’s evolving defence and security policies aim to close legal gaps exposed by modern threats such as cyberattacks and sabotage of undersea cables.

Plans for a vast AI data hub in the UAE have raised security concerns in Washington due to the country’s close ties with China.

LEGAL

Denmark’s Ministry of Culture has introduced a draft law aimed at safeguarding citizens’ images and voices under national copyright legislation.

US President Donald Trump has announced a 90-day extension for TikTok’s Chinese parent company, ByteDance, to secure a US buyer, effectively postponing a nationwide ban of the popular video-sharing app.

OpenAI reportedly tries to reduce Microsoft’s exclusive control over hosting its AI models, signalling growing friction between the two companies.

A federal judge in New York ordered the US Office of Personnel Management (OPM) to stop sharing sensitive personal data with the Department of Government Efficiency (DOGE) agents.

The European Commission has imposed a €329 million fine on Berlin-based Delivery Hero and its Spanish subsidiary, Glovo, for participating in what it described as a cartel in the online food delivery market.

ECONOMY

OpenAI chief executive Sam Altman revealed that he had a conversation with Microsoft CEO Satya Nadella on Monday to discuss the future of their partnership.

SoftBank founder Masayoshi Son is planning what could become his most audacious venture yet: a $1 trillion AI and robotics industrial park in Arizona.

Bitcoin prices slumped in mid-June as geopolitical tensions in the Middle East worsened.

OKX has expanded its European presence by launching fully compliant centralised exchanges in Germany and Poland.

Taiwan has officially banned the export of chips and chiplets to China’s Huawei and SMIC, joining the US in tightening restrictions on advanced semiconductor transfers.

Amazon will invest AU$ 20 billion to expand its data centre infrastructure in Australia, using solar and wind power instead of traditional energy sources.

SECURITY

A ransomware attack on the Swiss non-profit Radix has led to the theft and online publication of sensitive government data.

The Iran–Israel conflict has now expanded into cyberspace, with rival hacker groups launching waves of politically driven attacks.

Chinese AI company DeepSeek is gaining traction in global markets despite growing concerns about national security.

The National Cyber Security Centre (NCSC) has published new guidance to assist organisations in meeting the upcoming EU Network and Information Security Directive (NIS2) requirements.

A coalition of cybersecurity agencies, including the NSA, FBI, and CISA, has issued joint guidance to help organisations protect AI systems from emerging data security threats.

In a bold move highlighting growing concerns over digital sovereignty, the German state of Schleswig-Holstein is cutting ties with Microsoft.

Prior to the Danish government’s formal decision, the cities of Copenhagen and Aarhus had already announced plans to reduce reliance on Microsoft software and cloud services.

The EU Council and European Parliament have reached a political agreement to strengthen cross-border enforcement of the General Data Protection Regulation (GDPR).

One of the largest-ever leaks of stolen login data has come to light, exposing more than 16 billion records across widely used services, including Facebook, Google, Telegram, and GitHub.

A cyberattack has reportedly hit Iran’s Bank Sepah by the hacktivist group Predatory Sparrow.

Over 20,000 malicious IP addresses and domains linked to data-stealing malware have been taken down during Operation Secure, a coordinated cybercrime crackdown led by INTERPOL between January and April 2025.

The FBI has issued a warning about the resurgence of BADBOX 2.0, a dangerous form of malware infecting millions of consumer electronics globally.

US President Donald J. Trump signed a new Executive Order (EO) aimed at amending existing federal cybersecurity policies.

Japan’s parliament has passed a new law enabling active cyberdefence measures, allowing authorities to legally monitor communications data during peacetime and neutralise foreign servers if cyberattacks occur.

DEVELOPMENT

NATO is discussing proposals to broaden the scope of defence-related expenditures to help member states meet a proposed spending target of 5% of GDP.

SOCIO-CULTURAL

Social media platform X has updated its developer agreement to prohibit the use of its content for training large language models.

The UK’s Financial Conduct Authority (FCA) has taken action against unauthorised financial influencers in a coordinated international crackdown, resulting in three arrests.

Wikipedia has paused a controversial trial of AI-generated article summaries following intense backlash from its community of volunteer editors.

TikTok has globally banned the hashtag ‘SkinnyTok’ after pressure from the French government, which accused the platform of promoting harmful eating habits among young users.

For more information on cybersecurity, digital policies, AI governance and other related topics, visit diplomacy.edu.

EU Digital Diplomacy: Geopolitical shift from focus on values to economic security

The new EU International Digital Strategy 2025 (published on 5 June 2025) pivots from the EU’s values-based digital diplomacy (outlined in the 2023 Council Conclusions on Digital Diplomacy) towards a more geopolitical, security – and competition–driven approach.

The strategy is comprehensive in coverage, from digital infrastructure and security to economic issues and human rights. Yet, in a wide range of policy initiatives, it prioritises issues such as secure digital infrastructure around submarine cables and gives less prominence to, for example, the use of the ‘Brussels effect’ of spreading the EU’s digital regulations worldwide.

The strategy provides a multi-tier approach to the EU’s digital relations with the world, starting from global negotiations (the UN, G7, G20) via engagement with regional organisations (e.g. African Union, ASEAN) and specific countries.

Lastly, the strategy leaves many questions open, especially regarding its practical implementation.

This analysis dives deep into these and other aspects of the new EU international digital strategy, which will have a broader impact on tech developments worldwide. It compares the 2023 Council Conclusions on Digital Diplomacy with the 2025 EU International Digital Strategy. Both documents address the European Union’s international digital policy.

However, they differ significantly in form and drafting process. The 2023 Conclusions on EU Digital Diplomacy (The 2023 Conclusions) were adopted by the Council of the EU following negotiations among Member States. Conversely, the 2025 EU International Digital Strategy (The 2025 Strategy) was drafted by the European Commission and presented as a Communication to the Council and the European Parliament.

1. Key shifts in the 2025 Strategy

This brief unpacks these shifts, contrasting 2025’s priorities with the 2023 outlook, and concludes with policy recommendations.

From values to geopolitics and geoeconomics

The 2023 Conclusions stressed a human-centric, rights-based framework. The 2025 Strategy frames digital issues chiefly as matters of systemic resilience, economic competition, and security. The EU will focus more on building trade and security partnerships for exporting the EU’s AI and digital solutions than on ‘exporting’ norms (‘Brussels effect’).

The Strategy repeatedly stresses that boosting EU technological capacity (AI, semiconductors, cloud, quantum) is essential for economic growth and security. Tech Commissioner Virkkunen has stated that ‘tech competitiveness is an economic and security imperative’ for Europe.

Continued values, but subsumed

Although the Strategy still affirms support for human rights, it treats ‘values’ as subsidiary to strategic goals. In practice, issues like privacy and inclusivity are mentioned mostly in passing. By contrast, the 2023 Council Conclusions had foregrounded a ‘human-centric regulatory framework for an inclusive digital transformation’. The 2025 text incorporates those values under broader objectives of resilience and competitiveness.

Regulatory power de-emphasised

The EU’s traditional approach of using single‑market rules to set global standards (‘Brussels effect’) is notably downplayed. For example, the new strategy does not mention the AI Act – a flagship EU regulation. Instead, emphasis is on investment and cooperation (e.g. AI infrastructure) rather than the export of EU rules. This suggests the EU pragmatically recognises limits to unilateral rule-setting and focuses on building capabilities instead.

Team Europe approach

Both the 2023 Council Conclusions and the 2025 Strategy stress a coordinated EU approach. The Strategy prioritises deepening existing Digital Partnerships and Dialogues, and establishing new ones. A new Digital Partnership Network is proposed to coordinate these efforts, signalling an organisational shift toward structured cooperation.

Table 1: Key shifts in EU digital diplomacy (2023 vs. 2025)

2. New frontiers: What is the EU prioritising?

The 2025 Strategy introduces several priorities that were either absent or significantly less prominent in the 2023 Conclusions. These new focus areas illustrate the concrete manifestation of the EU’s intensified geopolitical and geoeconomic orientation.

Defence-linked technologies

For the first time, the EU links advanced digital tech to defence. It calls for efforts in cyber-defence, secure supply chains, and countering hybrid threats alongside AI, chips, and quantum R&D. The aim is an industry able to design and produce strategic tech (AI, semiconductors, cloud, quantum) at scale for both civilian and defence use.

This represents a significant securitisation of digital policy, moving beyond traditional cybersecurity to integrate technology directly into defence doctrines and industrial policy.

Secure infrastructure and connectivity

The strategy highlights investments in secure networks – 5G/6G, undersea (submarine) cables, and satellite links. Notably, it builds on the Global Gateway initiative (the EU’s alternative to China’s Belt and Road), co‑funding a network of secure submarine cables (Arctic, BELLA, MEDUSA, Blue-Raman), creating physical links with strategic partners and Digital Public Infrastructure in Africa, Asia, and Latin America. This addresses resilience against disruptions and foreign dependence in critical infrastructure.

Economic security – Trusted supply chains

The EU emphasises ‘resilient ICT supply chains’ and the use of trusted suppliers. In practice, this means diversifying away from over‑reliance on any one country or firm. The strategy also pushes digital trade frameworks: expanding digital trade agreements (e.g. with Singapore and South Korea) and promoting innovation in cooperation with ‘trusted partners’ to bolster EU leadership in emerging tech.

Digital Public Infrastructure (DPI)

There is a new focus on promoting EU-style DPIs abroad – for example, supporting partner countries to adopt secure digital IDs, e-services, data governance models, etc. The strategy calls for coordinated EU public-private investment in DPI and cybersecurity tools to aid partners’ digital transitions.

EU Tech Business Offer

This is a major new initiative – a public-private investment package (‘Tech Team Europe’) to help partners build digital capacity. Components include AI Factories (regional supercomputing/data centres), secure connectivity projects, digital skills and cyber capacity-building. The Strategy promises to roll out this dedicated Tech Business Offer globally, blending EU and member state resources to empower foreign markets with European tech (see Strategic Signals below).

Deepening partnerships and standards

The Strategy commits to expanding the Digital Partnerships and Dialogues established in 2023, and creating a new Digital Partnership Network to coordinate them. This means more joint R&D programmes (e.g., in quantum and semiconductors with Japan, Canada, and South Korea) and pushing for interoperable standards. The EU will continue to promote a rules-based digital order in line with its values, but through collaboration rather than just unilateral rules.

3. Shifting priorities: What’s less emphasised

While new priorities emerge, some traditional EU themes recede in prominence within the 2025 Strategy, though most are still present, with less emphasis. This highlights strategic trade-offs and potential implications of these shifts.

Brussels effect

The strategy downplays the ‘Brussels effect’, the EU’s reliance on its regulatory power to shape global norms. For example, the General Data Protection Regulation (GDPR) is mentioned only in passing, and the landmark AI Act is entirely absent. The Digital Services Act (DSA) is primarily discussed in an internal EU context, with little ambition to project its influence globally.

This signals a shift in the EU’s approach to ‘regulatory diplomacy,’ reflecting a broader reassessment of its approach to digital regulations. At the 2025 Paris AI Summit, both Ursula von der Leyen and Emmanuel Macron signalled a move toward slowing down AI and digital regulation.

This change responds to a wide range of criticisms: that the EU may have moved too far and too fast with the AI Act (see next section); that its regulatory approach is creating friction with the Trump administration, which views such measures as a threat to the US tech sector; and internal concerns that an overemphasis on regulation could hinder the EU’s competitiveness in the AI race against China and the United States.

EU AI Act

The marginalisation of the EU AI Act in the strategy reflects a broader shift within the EU. Although the Act was negotiated and adopted during a ‘public tsunami’ of concern over long-term AI risks, other actors have since retreated from this approach.

Examples include the slowing down of the Bletchley AI initiative during the Paris AI Summit and the Trump administration’s recalibrated AI strategy. In contrast, the EU faces a unique challenge: its approach to the regulation of AI models, heavily influenced by long-termist thinking, is already codified into law through the AI Act.

The primary challenge now lies in implementing the AI Act. The Act’s provisions related to the top two layers of the AI regulatory pyramid (see image of AI governance pyramid) have been gaining in relevance: protecting data and knowledge from misuse by AI platforms, and safeguarding human rights, consumer interests, employment, and education from adverse AI impacts.

The image shows a diagram of a four layered pyramid depicting AI governance. The bottom layer is labelled Hardware: AI computation. The second layer is labelled Algorithm, AI capabilities. The third layer is labelled Data & knowledge, AI sources. The tope of the pyramid is lavelled Agents, AI uses.

However, the provisions for regulating algorithms and long-term risks face practical and conceptual challenges. Some of these provisions, such as identifying the power of AI models mainly with the quantitative parameters of the number of FLOPs, are already outdated due to rapid technological advancements. There are criticisms that regulating algorithms may impact the competitiveness of the EU’s emerging AI industry.

Values-first framing

‘Human‑centric’ language still appears, but under resilience. Explicit human rights advocacy, such as protections for dissidents online or campaigns against censorship and surveillance, is barely mentioned in the strategy.In contrast, the 2023 Conclusions devoted numerous paragraphs to ‘vulnerable… groups’ (women, children, the disabled) and digital literacy.

The 2025 Strategy only briefly nods to ‘fundamental values’ in passing (e.g., promoting a rules-based order). Notably, it omits references to ‘internet shutdowns, online censorship and unlawful surveillance’, which were part of the 2023 document.

Standalone cyber diplomacy

The Strategy blurs the line between digital and cyber policy. Whereas in the past, the EU approaches treated cybersecurity and digital cooperation as distinct tracks, the new document integrates them. (As one example, the traditional ‘cyber diplomacy toolbox’ is now subsumed under broad tech partnerships.)

Focus on trade agreements

E-commerce policy has also shifted from a detailed 2023 focus on WTO e-commerce negotiations and the moratorium on customs duties towards highlighting the importance of bilateral/regional trade agreements as a primary tool for digital governance.

Table 2: Summary of prioritised and deprioritised issues in the 2025 Strategy

Prioritised issues

– Dual-use technologies & defence (AI, semiconductors, quantum linked to security)
– Submarine cables (resilience, cybersecurity focus)
– Economic & supply-chain security (resilient ICT, trusted suppliers)
– Tech competitiveness via trade & innovation (digital trade agreements)
– Digital infrastructure (5G/6G, DPI, AI factories, international cooperation)
– ‘EU Tech Business Offer’ (public-private investment abroad)

Deprioritised issues

– ‘Brussels effect’ (downplaying regulatory power, GDPR, AI Act, DSA)
– Explicit human rights advocacy (protections for dissidents, anti-censorship campaigns)
– E-commerce negotiations (shift from focus on WTO solutions towards bilateral/regional agreements)
– Prominence of global digital policies (e.g. Global Digital Compact, WSIS)

4. Geography of the EU’s digital diplomacy

The Strategy proposes multi-layered engagement combining bilateral arrangements (e.g. digital partnerships), regional framework (e.g. EU-LAC alliance) and global governance (GDC and WSIS) approach from the global via regional to national levels as summarised below.

Global level

The Strategy highlights the following initiatives and policy processes:

G7/G20/OECD: Coordination on AI safety, economic security standards, and semiconductors
UN: Implementation of Global Digital Compact (GDC), WSIS+20 Review
ITU: Rules-based radiofrequency allocation
Counter Ransomware Initiative: Joint operations against cybercrime
Clean Energy Ministerial: AI-energy collaboration

Regional level

Africa

– AI Hub for Sustainable Development (co-designed with UNDP)
– Collaboration with Smart Africa (Africa AI Council) under Global Gateway
– ‘AI for Public Good’ initiative (Generative AI solutions, capacity building)

Cybersecurity

– Cyber capacity-building projects (strategic frameworks, incident management)

ASEAN

Security connectivity

– Expansion of Copernicus mirror site to the Philippines, Indonesia, Thailand (EU-ASEAN Sustainable Connectivity Initiative)

Global digital governance

– Regional events on human rights-based platform governance

Central America (SICA)

Digital identities/DPI

– Mutual recognition of e-signatures (Costa Rica, El Salvador, Guatemala, Honduras, Nicaragua, Panama)

Latin America and the Caribbean

Security connectivity

– BELLA submarine cable (EU-LAC)
– LAC Connectivity Toolbox development
– MEDUSA cable extension to West Africa

Quantum/HPC

EU-LAC Supercomputing Network (federated HPC resources)

Cybersecurity

– LAC4 cybersecurity education hub
– EU-LAC Cybersecurity Community of Practice

‘AI for Public Good’ initiative

Western Balkans

There is ambiguity in dealing with the Western Balkans. While the regional approach is used in the EU’s policies, including this strategy, the EU engages countries from the region individually.

Security connectivity

– Alignment with the EU 5G Cybersecurity Toolbox

Digital identities/DPI

– Preparation for integration with the EU Digital Identity Wallet
– Mutual recognition of e-signatures
– Onboarding to Single Digital Gateway

Online platforms

– DSA/DMA alignment support

Country level

The Strategy’s linguistic statistics show India and Japan’s growing relevance for the EU’s digital diplomacy.

Table 3: Frequency of country referencing in 2025 Strategy

India: Strategic partnership

The relevance of India for the EU’s digital diplomacy has risen significantly. It reflects the overall EU geopolitical shift towards India as a democracy, a leading tech actor, and a country with the highest strategic capacity to connect various groups and blocs, including BRICS, G20, and G77.

EU-India cooperation initiatives

Security connectivity

– Blue-Raman submarine cable (via India-Middle East-Europe Economic Corridor)

Semiconductors

– Talent exchange programme for semiconductor skills (under EU-India TTC)

Cybersecurity

– Cyber Dialogues
– Exploration of Mutual Recognition Agreements (Cyber Resilience Act)

Digital identities/DPI

– Cooperation on e-signatures and DPI interoperability
– Promotion of EU eID Wallet model

– Cooperation on e-signatures and DPI interoperability
– Promotion of the EU eID Wallet model

Like-minded partners: Deepening alliances

The EU explicitly seeks deeper digital ties with like‑minded countries, including Japan, South Korea, Canada, India, Singapore, and the United Kingdom. A first Digital Partnership Network meeting (involving EU and partner states) is planned to coordinate tech cooperation. Joint research initiatives are slated (quantum, semiconductor programs with Japan, Canada, South Korea).

Australia

Cybersecurity

– Cyber Dialogues: Exploration of Mutual Recognition Agreements (Cyber Resilience Act)

– Administrative arrangement on AI safety (via AI Safety Institutes network)

Canada

Security connectivity

– Arctic submarine cable projects (with EU/US/Japan)

Quantum/HPC

– Joint quantum research projects (computing, sensing, communication)

– AI safety cooperation (via AI Safety Institutes network)

Japan

Security connectivity

– Arctic submarine cable projects (with EU/US/Canada)

Quantum/HPC

– Joint quantum-HPC research (biomedical/environmental sciences)

Semiconductors

Joint R&D projects (PFAS replacement, chiplets, sub-2nm processes)

Cybersecurity

– Cyber Dialogues
– Exploration of Mutual Recognition Agreements (Cyber Resilience Act)

Digital identities/DPI

– Mutual recognition of e-signatures
– eID Wallet interoperability cooperation

– AI safety cooperation (via AI Safety Institutes network)
– Joint research on AI innovation

United Kingdom

Cybersecurity

– Cyber Dialogues

– AI safety cooperation (via AI Safety Institutes network)

Norway

Defence innovation

– Synergies with Hub for European Defence Innovation (HEDI)

Republic of Korea

Semiconductors

– Joint R&D projects (heterogeneous chip integration, neuromorphic computing)

Quantum/HPC

– Joint quantum technology projects (computing, sensing, communication)

5G/6G

– 6G R&D cooperation (AI-powered RAN)

Cybersecurity

– Cyber Dialogues

– AI safety cooperation (via AI Safety Institutes network)

Singapore

Cybersecurity

-Promotion of the EU eID Wallet model

Digital identities/DPI

– Promotion of the EU eID Wallet model

Taiwan

5G/6G

– Potential 6G research cooperation (physical layers, antennas)

Semiconductors

– Participation in the World Semiconductor Council (GAMS)

United States: Continue, wait and see

The uncertainty of EU-US digital relations is signalled by the lack of references in the strategy of the EU-US Trade Technology Council, which was the main mechanism for transatlantic digital cooperation. However, the EU signalled its readiness to cooperate, emphasising that it ‘remains a reliable and predictable partner’.

Several ‘omissions’ in the strategy also underscore this readiness to engage. Notably, tech companies are not singled out for criticism, as they often are in the EU’s tech sovereignty approaches and numerous anti-monopoly and content policy initiatives. There is no explicit reference to digital services taxes for tech companies, which has historically been a strong card in the EU’s geoeconomic relations with the US.

EU-USA cooperation initiatives

Security connectivity

– Arctic submarine cable projects (with EU/Canada/Japan)

Semiconductors

– Joint early warning mechanism for supply chains
– Transparency mechanism on semiconductor subsidies

Cybersecurity

– Cyber Dialogues

China: Between cooperation and competition

China is scarcely mentioned in the text – digital cooperation is deferred to an upcoming summit. However, the Strategy is clearly defensive: it doubles down on ‘secure and trusted 5G networks’ globally, implicitly excluding Chinese vendors like Huawei. It also casts the EU’s Global Gateway as a digital alternative to China’s Belt and Road, investing in secure cables and AI infrastructure in Asia, Africa and Latin America. In sum, the signal to Beijing is one of wary competition. Digital relations will be discussed during the forthcoming China-EU Summit in July 2025.

Neighbourhood: Deep engagement

Regionally, the EU will push its Digital Single Market model to its neighbours. For instance, Ukraine, Moldova, and the Western Balkans are targeted to integrate EU digital rules rapidly (secure IDs, connectivity, regulatory alignment).

Moldova

Security connectivity

– Alignment with EU 5G Cybersecurity Toolbox

Digital identities/DPI

– Preparation for integration with EU Digital Identity Wallet
– Extension of EU Cyber-Reserve

Cybersecurity

– Cyber capacity building

Ukraine

Security connectivity

– Development of Black Sea digital links

Cybersecurity

– Cyber Dialogues
– Extension of the EU Cyber-Reserve

Digital identities/DPI

– Preparation for integration with the EU Digital Identity Wallet
– AI-based Local Digital Twins for urban reconstruction

Defence innovation

– Enhanced collaboration via the EU Defence Innovation Scheme (EUDIS)
– Synergies with Hub for European Defence Innovation (HEDI)

Global South: Global inclusion

The EU plans to expand Global Gateway digital projects for Africa, Asia, and Latin America: co-financing secure submarine cables, undersea connectivity to Europe, and building local digital infrastructure. A dedicated Tech Business Offer will extend to the Southern Neighbourhood and sub-Saharan Africa.

Brazil

Cybersecurity

– Cyber Dialogues

Digital identities/DPI

– Cooperation on e-signatures and DPI interoperability

Egypt

Digital identities/DPI

– Mutual recognition of trust services
– eID Wallet interoperability cooperation

Costa Rica

5G/6G

– 5G testbed partnership (with Nokia/Ericsson)

5. Open questions

Several critical questions will impact the implementation of the Strategy.

Synchronising internal and external digital policy

So far, the EU’s internal and foreign policies have been in sync. Europe has been practising locally what was preached abroad. The new Strategy may change these dynamics as Brussels remains committed to a values-based domestic order. Still, externally, the Strategy recasts digital policy in terms of global rivalry and resilience in security and economic domains. This shift from principles to geopolitical interests will trigger tensions in the implementation, as the two approaches necessitate different policy instruments, methods, and language.

Defining digital and internet governance

The Strategy repeatedly speaks of ‘digital’ affairs, but it also includes several references to ‘internet governance’, leaving it unclear how the EU sees the two concepts and the distinctions between them (if any). In practice, nearly all tech issues (from infrastructure to AI) fall under the internet’s umbrella, so perhaps a clear separation between the two is not very straightforward. Clarifying what means what for the EU is needed to avoid confusion over overlapping mandates.

Cyber vs. digital diplomacy coordination

The EU has two separate diplomatic tracks for cybersecurity and overall digital issues. The new Strategy implies that these converge, but the organisational plan is vague. Will the EU continue to have two – cyber and digital – ambassador networks, or will there be an integrated structure? The cyber-digital coordination will rise in importance as the strategy is implemented. The forthcoming Danish presidency of the EU in the second part of 2025 may provide some solutions as Denmark has experience in running holistic tech diplomacy, combining security, economic, and other policy aspects.

Implementation and resources

The Strategy misses implementation details. Which EU body will lead its rollout? How will funding be allocated across the many initiatives (Tech Business Offer, infrastructure projects, partnerships)? Importantly, will the EU follow up on the 2023 Conclusions, which explicitly call to ensure that ‘at least one official in every EU Delegation has relevant expertise on digital diplomacy‘ and that diplomats receive training?

Progress on this front will be a key test of delivery.

AI Strategy – Computing power vs data and knowledge

The Strategy heavily emphasises computing infrastructure (AI ‘gigafactories’). It follows global inertia but not AI research, which shows that adding GPU power does not significantly improve AI inference quality (diminishing returns).¹ Should the EU instead leverage its data strengths, broad knowledge pool, and human expertise? Ensuring quality data, skilled human capital, and innovative applications may be as important as raw computing. Balancing the EU’s high-education workforce and data resources with these new AI investments is an open policy debate.

6. Conclusion and recommendations

The 2025 Strategy marks a realpolitik turn for EU digital policy. It recognises that digital affairs are a geopolitical competition and an economic power domain. The EU is right to bolster its tech capabilities, diversify alliances, and secure critical infrastructure. However, it must not abandon its foundational advantages.

Keep values and ethics in view

Even as the strategy focuses on power and security, EU core values – human rights, privacy, democracy – remain the EU’s unique selling points. The EU should ensure that its external actions (e.g. tech exports, partnerships) reflect these values. For example, the Ethics Guidelines for AI or human rights due diligence in tech supply chains should accompany hard-power initiatives. Over time, this ‘soft’ aspect of tech diplomacy will underpin global trust in EU solutions, especially during the AI transformation of societies.

Leverage the knowledge ecosystem

Europe’s educated workforce, strong research base, and innovative companies are among its strengths. The Strategy should explicitly link new initiatives (like AI factories) to European data assets and talent. For instance, investing in common European data spaces, R&D hubs, and AI talent training will complement hardware investments. Policies should channel these resources into EU security and cooperative R&D projects with partners, maximising long-term benefits.

Building digital diplomacy capacity

The EU should develop digital diplomacy capabilities. The Commission and the High Representative for Foreign Affairs should clarify the future of the cyber/digital networks, ensuring diplomats and delegations have the skills to carry out the strategy. This could involve appointing Digital Envoys in key regions, scaling up EU-funded training programs for officials, and embedding digital attachés in trade missions. Progress should be reported regularly (as the 2023 Conclusions demanded).

Clarify governance frameworks

The EU should avoid confusion by clearly defining its terms. If ‘digital diplomacy’ now covers everything from 5G to AI, the Strategy (or follow-up guidelines) should explain how it relates to internet governance, cyber diplomacy, and other fields. This will help partners and stakeholders navigate the new agenda.

Continue global engagement

Finally, the EU should remember that its influence often came from combining hard and soft power. Alongside alliances, it should maintain active engagement in multilateral fora (e.g. UN Internet Governance Forum, WTO e-commerce activities) to promote interoperability and standards. The Strategy’s emphasis on bilateral ties should balance efforts to shape open, rules-based markets globally.

The 2025 Strategy lays out an ambitious vision of a stronger, more assertive EU in global tech. To make it work, policymakers should marry this realpolitik turn with the EU’s enduring strengths – its values, expertise, and rule-of-law model – and put concrete implementation in place.

Annex: The strategy and conceptual approach to digital diplomacy

The 2025 strategy covers the impact of digital technology on the geopolitical environment for diplomacy, and topics on the diplomatic agenda, as per Diplo’s methodology for the impact of digitalisation on diplomacy.

The Strategy uses the term ‘digital diplomacy’ only once in the context of a policy topic to advance ‘our international priorities and to build partnerships’. It does not address the use of digital tools for the conduct of diplomacy (e.g. the use of social media). The EU has been using ‘digital diplomacy’ to cover geopolitical changes and new topics on the diplomatic agenda, as summarised here.

The image shows an infographic entitled mapping the interplay between digitalisation and diplomacy, covering the digital impact on geopolitics and diplomacy, digital policy topics on diplomatic agenda, and digital tools for diplomatic activities.

As Ilya Sutskever (OpenAI co-founder) stated: ‘The 2010s were the age of scaling; now we’re back in the age of discovery’ (LINK) ↩︎

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

Advancing Swiss AI Trinity: Zurich’s entrepreneurship, Geneva’s governance, and communal subsidiarity

Switzerland can chart a unique path in the global AI race by combining three strengths: Zurich’s innovative entrepreneurship, Geneva’s responsible governance, and communal enabling subsidiarity.

Swiss AI Trinity can advance beyond technology by reimagining a national social contract for the AI age, firmly grounded in Swiss values.

The country’s AI transformation unfolds as the tech world faces profound flux. Amid the US push for unfettered tech growth and escalating US-China rivalry resembling an AI arms race, the EU and nations worldwide seek their own strategies. Uncertainty grows alongside deepening concerns about AI’s impact, from jobs and the economy to education and media.

In this fluid environment, Switzerland has a rare opportunity to carve out a distinctive approach to cutting-edge AI development, anchored in subsidiarity, apprenticeship, and national traditions.

Time is short, but the timing is favourable for Switzerland:

AI is commoditised: Switzerland’s delayed start in developing LLMs is no longer a disadvantage. New LLMs emerge daily; open-source models enable easy retraining; AI agents can be built in minutes. Success now hinges on data, knowledge, and uses of AI, not just hardware or algorithms. The key is the nexus between artificial and human intelligence, where Switzerland can excel due to its robust educational and apprenticeship model.

Pushback on AI regulation: After a frenzied regulatory race last few years, a more balanced approach now prioritises short-term risks (jobs, education) over existential long-term threats. The Trump administration has already slowed down US regulatory momentum, while the EU re-evaluates parts of its AI Act concerning generative models. Switzerland’s prudent regulatory stance has become an advantage. New AI regulation may be adopted as technology matures and crystallise real policy problems that should be regulated.

Against this backdrop, Switzerland’s AI Trinity proposes a three-pronged strategy:

Zurich: A hub of private-sector innovation;
Geneva: A crucible for global governance and standardisation;
Swiss cantons, cities, and communities: Upholding subsidiarity to drive inclusive, bottom-up AI development.

Each pillar builds on existing strengths while addressing the urgent need to rethink business models, governance, and social contracts for the AI era.

Zurich: Supercharging business and innovation

While cutting-edge technology often seems concentrated in massive data centres and trillion-dollar corporations, DeepSeek exemplifies how breakthroughs can spring from lean, agile labs. Zurich is ideally placed to harness both approaches.

World-class ecosystem: ETH Zurich—ranked among the world’s top universities—alongside R&D hubs for Microsoft, Google, and others, provides the talent, research excellence, and entrepreneurial mindset to keep Switzerland at AI’s forefront.
Global reach, diverse perspectives: Partnerships must extend beyond Silicon Valley. Engaging Chinese and Indian tech players fosters competition, sparks creativity, and mitigates over-reliance on Western supply chains.
Stronger academic-industry ties: Deeper collaboration between ETH and private-sector leaders will accelerate ventures. Breakthroughs in healthcare, climate tech, and beyond emerge when elite research meets real-world application.

Zurich is thus not just a global financial centre, but a beacon for responsible, human-centric AI.

Geneva: Forging global governance and standards

While Zurich drives innovation, Geneva can shape balanced global AI governance—a race intensifying amid rival initiatives from the Gulf and emerging tech hubs. Geneva must act decisively:

Translating hype into action: Nations urgently need pragmatic AI tools for pandemic response, environmental risk, and equitable education. Geneva-based bodies should prioritise tangible solutions, dispelling perceptions of AI as abstract hype.
Mainstreaming AI: Treating AI as integral to health, commerce, and labour rights is vital. As AI becomes a universal necessity, Geneva’s institutions must weave digital policy into all major global negotiations.
Modernising international organisations: Outdated, top-heavy management of international organisations lacks the agility for rapid AI evolution. Embedding AI translation services, automated reporting, and similar tools will boost transparency, efficiency, and collaboration.
Defining standards where treaties fall short: Without robust international agreements, technical standards ensure interoperability. Geneva’s standardisation expertise positions it to lead in healthcare AI, digital trade, and environmental data.

By embracing this role, Geneva can ensure that multilateral bodies guide AI ethics in a transforming world.

Communities and cantons: Inclusion through subsidiarity

Switzerland’s greatest AI advantage lies in subsidiarity—the principle of localised decision-making. Distributing AI development across cantons and communities ensures innovation aligns with real needs, addresses local contexts, and leaves no one behind. Key activities include:

‘AI for All’ programme: Offer small grants to citizens and small businesses for developing AI agents, democratising access to tools and catalysing solutions rooted in local needs.
AI education and apprenticeship: Develop AI apprenticeship, building on the Swiss long tradition of vocational training; integrate AI education from primary schools to universities.
Libraries and local AI labs: Repurpose libraries, community centres, and post offices into AI knowledge hubs. Applying machine learning to hyper-local challenges – agriculture, tourism, health – can empower communities as active creators of AI innovation.

A call to action: Switzerland’s moment to lead

Switzerland stands at a decisive juncture. Uniting Zurich, Geneva, and its cantons can nurture an AI future that is cutting-edge yet fair, transparent, and Swiss at its core.

By adopting this AI Trinity approach—balancing innovation, governance, and subsidiarity—Switzerland can show the world how to embrace advanced technology without sacrificing societal values.

Practical steps forward:

Scale AI apprenticeships through Switzerland’s vocational tradition.
Launch a national AI capacity-building programme for citizens and companies.
Adapt school and university curricula to foster creativity as AI automates tasks like essay drafting.
Repurpose libraries and post offices into community knowledge hubs.
Prioritise ‘AI for All’ principles in public projects, procurement, and grants at all governmental levels.

Swiss AI Trinity approach can democratise AI, empower local innovation, and fuel inclusive growth from the ground up. The tools, talent, and tradition are in place. The time to act is now.

For more information on these topics, visit diplomacy.edu.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

Breaking down the OEWG’s legacy: Hits, misses, and unfinished business

The OEWG on cybersecurity (2019–2025) shaped global debates on digital security, but did it deliver? External experts weigh in on its lasting impact, while our team, who tracked the process from day one, dissect the milestones and missed opportunities. Together, these perspectives reveal what’s next for cyber governance in a fractured world.

What is the OEWG?

The open-ended working groups (OEWGs) are a type of format present in the UN that is typically considered the most open, as the name suggests. It means that all UN member and observer states, intergovernmental organisations, and non-governmental organisations with the UN Economic and Social Council (ECOSOC) consultative status may attend public meetings of the working group. Yet, decisions are made by the UN member states. There are various OEWGs at the UN. Here, we are addressing the one dealing with cybersecurity.

What does the OEWG on cybersecurity do? In plain language, it tries to find more common ground on what is allowed and what is not in cyberspace, and how to ensure adherence to these rules. In the UN language, the Cyber OEWG was mandated to ‘continue to develop the rules, norms, and principles of responsible behaviour of states, discuss ways for their implementation, and to study the possibility of establishing regular institutional dialogue with broad participation under the auspices of the UN.’

How was the OEWG organised? The OEWG was organised around an organisational session that discussed procedures and modus operandi, and substantive ones dealing with the matter, as well as intersessional meetings and town halls supplementing the discussions. The OEWG held 10 substantive sessions during its 5-year mandate, with the 11th and final session just around the corner in July 2025, where the group will adopt its Final report.

The OEWG through expert eyes: Achievements, shortfalls, and future goals

As the OEWG 2019–2025 process nears its conclusion, we spoke with cybersecurity experts to reflect on its impact and look ahead. Their insights address four key questions:

The OEWG’s most substantive contributions and shortcomings in global ICT security
Priorities for future dialogues on responsible state behaviour in cyberspace
The feasibility of consensus on a permanent multilateral mechanism
The potential relevance of such a mechanism in today’s divisive geopolitical climate

Their perspectives shed light on what the OEWG has achieved—and the challenges still facing international cyber governance.

In addition to external cybersecurity experts, we asked our own team—who have tracked the OEWG process since its inception—to share their analysis. They highlight key achievements over the past five years, identify gaps in the discussions, and offer predictions on where debates may lead during the final session and beyond.

Read the full expert commentaries on our dedicated web page.

Meme coins: Fast gains or crypto gambling?

A high-stakes game of digital chance, meme coins blur the line between viral entertainment and financial risk in the wildest corner of crypto.

Meme coins have exploded in the crypto market, attracting investors with promises of fast profits and viral hype. These digital tokens, often inspired by internet memes and pop culture, like Dogecoin, Pepe, Dogwifhat and most recently Trump coin, do not usually offer clear utility. Instead, their value mostly depends on social media buzz, influencer endorsements, and community enthusiasm. In 2025, meme coins remain a controversial yet dominant trend in crypto trading.

Viral but vulnerable: the rise of meme coins

Meme coins are typically created for humour, social engagement, or to ride viral internet trends, rather than to solve real-world problems. Despite this, they are widely known for their popularity and massive online appeal. Many investors are drawn to meme coins because of the potential for quick, large returns.

For example, Trump-themed meme coins saw explosive growth in early 2024, with MAGA meme coin (TRUMP) briefly surpassing a $500 million market cap, despite offering no real utility and being driven largely by political hype and social media buzz.

Analysis reports indicate that in 2024, between 40,000 and 50,000 new meme tokens were launched daily, with numbers soaring to 100,000 during viral surges. Solana tops the list of blockchains for meme coin activity, generating 17,000 to 20,000 new tokens each day.

Chainplay’s ‘State of Memecoin 2024’ report found that over half (55.24%) of the meme coins analysed were classified as ‘malicious’.

The risks of rug pulls and scams in meme coin projects

Beneath the humour and viral appeal, meme coins often hide serious structural risks. Many are launched by developers with little to no accountability, and most operate with centralised liquidity pools controlled by a small number of wallets. The setup allows creators or early holders to pull liquidity or dump large token amounts without warning, leading to devastating price crashes—commonly referred to as ‘rug pulls.’

On-chain data regularly reveals that a handful of wallets control the vast majority of supply in newly launched meme tokens, making market manipulation easy and trust almost impossible. These coins are rarely audited, lack transparency, and often have no clear roadmap or long-term utility, which leaves retail investors highly exposed.

The combination of hype-driven demand and opaque tokenomics makes meme coins a fertile ground for fraud and manipulation, further eroding public confidence in the broader crypto ecosystem.

Gambling disguised as investing: The adrenaline rush of meme coins

Meme coins tap into a mindset that closely resembles gambling more than traditional investing. The entire culture around them thrives on adrenaline-fueled speculation, where every price spike feels like hitting a jackpot and every drop triggers a high-stakes rollercoaster of emotions. Known as the ‘degen’ culture, traders chase quick wins fuelled by FOMO, hype, and the explosive reach of social media.

The thrill-seeking mentality turns meme coin trading into a game of chance. Investors often make impulsive decisions based on hype rather than fundamentals, hoping to catch a sudden pump before the inevitable crash.

It is all about momentum. The volatile swings create an addictive cycle: the excitement of rapid gains pulls traders back in, despite the constant risk of losing everything.

While early insiders and large holders strategically time their moves to cash out big, most retail investors face losses, much like gamblers betting in a casino. The meme coin market, therefore, functions less like a stable investment arena and more like a high-risk gambling environment where luck and timing often outweigh knowledge and strategy.

Is profit from meme coins possible? Yes, but…

While some investors have made substantial profits from meme coins, success requires expert knowledge, thorough research, and timing. Analysing tokenomics, community growth, and on-chain data is essential before investing. Although they can be entertaining, investing in meme coins is a risky gamble. Luck remains a big key factor, so meme coins are never considered safe or long-term investments.

Meme coins vs Bitcoin: A tale of two mindsets

Many people assume that all cryptocurrencies share the same mindset, but the truth is quite different. Interestingly, cryptocurrencies like Bitcoin and meme coins are based on contrasting philosophies and psychological drivers.

Bitcoin embodies a philosophy of trust through transparency, decentralisation, and long-term resilience. It appeals to those seeking stability, security, and a store of value rooted in technology and community consensus—a digital gold that invites patience and conviction. In essence, Bitcoin calls for building and holding with reason and foresight.

Meme coins, on the other hand, thrive on the psychology of instant gratification, social identity, and collective enthusiasm. They tap into our desire for excitement, quick wins, and belonging to a viral movement. Their value is less about utility and more about shared emotion— the hope, the hype, and the adrenaline rush of catching the next big wave. Meme coins beckon with the thrill of the moment, the gamble, and the social spectacle. It makes meme coins a reflection of the speculative and impulsive side of human nature, where the line between investing and gambling blurs.

Understanding these psychological underpinnings helps explain why the two coexist in the crypto world, yet appeal to vastly different types of investors and mindsets.

How meme coins affect the reputation of the entire crypto market

The rise and fall of meme coins do not just impact individual traders—they also cast a long shadow over the credibility of the entire crypto industry.

High-profile scams, rug pulls, and pump-and-dump schemes associated with meme tokens erode public confidence and validate sceptics’ concerns. Many retail traders enter the meme coin space with high hopes and are quickly disillusioned by manipulation and sudden losses.

This leads to a sense of betrayal, triggering risk aversion and a generalised mistrust toward all crypto assets, even those with strong fundamentals like Bitcoin or Ethereum. Such disillusionment does not stay contained. It spills over into mainstream sentiment, deterring new investors and slowing institutional adoption.

As more people associate crypto with gambling and scams rather than innovation and decentralisation, the market’s growth potential suffers. In this way, meme coins—though intended as jokes—could have serious consequences for the future of blockchain credibility.

Trading thrills or ticking time bomb?

Meme coins may offer flashes of fortune, but their deeper role in the crypto ecosystem raises a provocative question: are they reshaping finance or just distorting it? In a market where jokes move millions and speculation overrides substance, the real gamble may not just be financial—it could be philosophical.

Are we embracing innovation, or playing a dangerous game with digital dice? In the end, meme coins are not just a bet on price—they are a reflection of what kind of future we want to build in crypto. Is it sustainable value, or just viral chaos? The roulette wheel is still spinning.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

Cognitive offloading and the future of the mind in the AI age

The rise of AI is transforming work and education, but raises questions about its impact on critical thinking and cognitive independence.

AI reshapes work and learning

The rapid advancement of AI is bringing to light a range of emerging phenomena within contemporary human societies.

The integration of AI-driven tools into a broad spectrum of professional tasks has proven beneficial in many respects, particularly in terms of alleviating the cognitive and physical burdens traditionally placed on human labour.

By automating routine processes and enhancing decision-making capabilities, AI has the potential to significantly improve efficiency and productivity across various sectors.

In response to these accelerating technological changes, a growing number of nations are prioritising the integration of AI technologies into their education systems to ensure students are prepared for future societal and workforce transformations.

China advances AI education for youth

China has released two landmark policy documents aimed at integrating AI education systematically into the national curriculum for primary and secondary schools.

The initiative not only reflects the country’s long-term strategic vision for educational transformation but also seeks to position China at the forefront of global AI literacy and talent development.

The two guidelines, formally titled the Guidelines for AI General Education in Primary and Secondary Schools and the Guidelines for the Use of Generative AI in Primary and Secondary Schools, represent a scientific and systemic approach to cultivating AI competencies among school-aged children.

Their release marks a milestone in the development of a tiered, progressive AI education system, with carefully delineated age-appropriate objectives and ethical safeguards for both students and educators.

The USA expands AI learning in schools

In April, the US government outlined a structured national policy to integrate AI literacy into every stage of the education system.

By creating a dedicated federal task force, the administration intends to coordinate efforts across departments to promote early and equitable access to AI education.

Instead of isolating AI instruction within specialised fields, the initiative seeks to embed AI concepts across all learning pathways—from primary education to lifelong learning.

The plan includes the creation of a nationwide AI challenge to inspire innovation among students and educators, showcasing how AI can address real-world problems.

The policy also prioritises training teachers to understand and use AI tools, instead of relying solely on traditional teaching methods. It supports professional development so educators can incorporate AI into their lessons and reduce administrative burdens.

The strategy encourages public-private partnerships, using industry expertise and existing federal resources to make AI teaching materials widely accessible.

European Commission supports safe AI use

As AI becomes more common in classrooms around the globe, educators must understand not only how to use it effectively but also how to apply it ethically.

Rather than introducing AI tools without guidance or reflection, the European Commission has provided ethical guidelines to help teachers use AI and data responsibly in education.

Published in 2022 and developed with input from educators and AI experts, the EU guidelines are intended primarily for primary and secondary teachers who have little or no prior experience with AI.

Instead of focusing on technical complexity, the guidelines aim to raise awareness about how AI can support teaching and learning, highlight the risks involved, and promote ethical decision-making.

The guidelines explain how AI can be used in schools, encourage safe and informed use by both teachers and students, and help educators consider the ethical foundations of any digital tools they adopt.

Rather than relying on unexamined technology, they support thoughtful implementation by offering practical questions and advice for adapting AI to various educational goals.

AI tools may undermine human thinking

However, technological augmentation is not without drawbacks. Concerns have been raised regarding the potential for job displacement, increased dependency on digital systems, and the gradual erosion of certain human skills.

As such, while AI offers promising opportunities for enhancing the modern workplace, it simultaneously introduces complex challenges that must be critically examined and responsibly addressed.

One significant challenge that must be addressed in the context of increasing reliance on AI is the phenomenon known as cognitive offloading. But what exactly does this term entail?

What happens when we offload thinking?

Cognitive offlo ading refers to the practice of using physical actions or external tools to modify the information processing demands of a task, with the aim of reducing the cognitive load on an individual.

In essence, it involves transferring certain mental functions—such as memory, calculation, or decision-making—to outside resources like digital devices, written notes, or structured frameworks.

While this strategy can enhance efficiency and performance, it also raises concerns about long-term cognitive development, dependency on technological aids, and the potential degradation of innate mental capacities.

How AI may be weakening critical thinking

A study, led by Dr Michael Gerlich, Head of the Centre for Strategic Corporate Foresight and Sustainability at SBS Swiss Business School, published in the journal Societies raises serious concerns about the cognitive consequences of AI augmentation in various aspects of life.

The study suggests that frequent use of AI tools may be weakening individuals’ capacity for critical thinking, a skill considered fundamental to independent reasoning, problem-solving, and informed decision-making.

More specifically, Dr Gerlich adopted a mixed-methods approach, combining quantitative survey data from 666 participants with qualitative interviews involving 50 individuals.

Participants were drawn from diverse age groups and educational backgrounds and were assessed on their frequency of AI tool use, their tendency to offload cognitive tasks, and their critical thinking performance.

The study employed both self-reported and performance-based measures of critical thinking, alongside statistical analyses and machine learning models, such as random forest regression, to identify key factors influencing cognitive performance.

Younger users, who rely more on AI, think less critically

The findings revealed a strong negative correlation between frequent AI use and critical thinking abilities. Individuals who reported heavy reliance on AI tools—whether for quick answers, summarised explanations, or algorithmic recommendations—scored lower on assessments of critical thinking.

The effect was particularly pronounced among younger users aged 17 to 25, who reported the highest levels of cognitive offloading and showed the weakest performance in critical thinking tasks.

In contrast, older participants (aged 46 and above) demonstrated stronger critical thinking skills and were less inclined to delegate mental effort to AI.

Higher education strengthens critical thinking

The data also indicated that educational attainment served as a protective factor: those with higher education levels consistently exhibited more robust critical thinking abilities, regardless of their AI usage levels.

These findings suggest that formal education may equip individuals with better tools for critically engaging with digital information rather than uncritically accepting AI-generated responses.

Now, we must understand that while the study does not establish direct causation, the strength of the correlations and the consistency across quantitative and qualitative data suggest that AI usage may indeed be contributing to a gradual decline in cognitive independence.

However, in his study, Gerlich also notes the possibility of reverse causality—individuals with weaker critical thinking skills may be more inclined to rely on AI tools in the first place.

Offloading also reduces information retention

While cognitive offloading can enhance immediate task performance, it often comes at the cost of reduced long-term memory retention, as other studies show.

The trade-off has been most prominently illustrated in experimental tasks such as the Pattern Copy Task, where participants tasked with reproducing a pattern typically choose to repeatedly refer to the original rather than commit it to memory.

Even when such behaviours introduce additional time or effort (e.g., physically moving between stations), the majority of participants opt to offload, suggesting a strong preference for minimising cognitive strain.

These findings underscore the human tendency to prioritise efficiency over internalisation, especially under conditions of high cognitive demand.

The tendency to offload raises crucial questions about the cognitive and educational consequences of extended reliance on external aids. On the one hand, offloading can free up mental resources, allowing individuals to focus on higher-order problem-solving or multitasking.

On the other hand, it may foster a kind of cognitive dependency, weakening internal memory traces and diminishing opportunities for deep engagement with information.

Within the framework, cognitive offloading is not a failure of memory or attention but a reconfiguration of cognitive architecture—a process that may be adaptive rather than detrimental.

However, the perspective remains controversial, especially in light of findings that frequent offloading can impair retention, transfer of learning, and critical thinking, as Gerlich’s study argues.

If students, for example, continually rely on digital devices to recall facts or solve problems, they may fail to develop the robust mental models necessary for flexible reasoning and conceptual understanding.

The mind may extend beyond the brain

The tension has also sparked debate among cognitive scientists and philosophers, particularly in light of the extended mind hypothesis.

Contrary to the traditional view that cognition is confined to the brain, the extended mind theory argues that cognitive processes often rely on, and are distributed across, tools, environments, and social structures.

As digital technologies become increasingly embedded in daily life, this hypothesis raises profound questions about human identity, cognition, and agency.

At the core of the extended mind thesis lies a deceptively simple question: Where does the mind stop, and the rest of the world begin?

Drawing an analogy to prosthetics—external objects that functionally become part of the body—Clark and Chalmers argue that cognitive tools such as notebooks, smartphones, and sketchpads can become integrated components of our mental system.

These tools do not merely support cognition; they constitute it when used in a seamless, functionally integrated manner. This conceptual shift has redefined thinking not as a brain-bound process but as a dynamic interaction between mind, body, and world.

Balancing AI and human intelligence

In conclusion, cognitive offloading represents a powerful mechanism of modern cognition, one that allows individuals to adapt to complex environments by distributing mental load.

However, its long-term effects on memory, learning, and problem-solving remain a subject of active investigation. Rather than treating offloading as inherently beneficial or harmful, future research and practice should seek to balance its use, leveraging its strengths while mitigating its costs.

Ultimately, we -as educators, policymakers, and technologists- have to shape the future of learning, work and confront a central tension: how to harness the benefits of AI without compromising the very faculties—critical thought, memory, and independent judgment—that define human intelligence.

For more information on these topics, visit diplomacy.edu.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

DW Weekly #219 – AGI or bust: Tech giants gamble big on the future of intelligence

27 June – 4 July 2025

Dear readers,

Over the past week, Meta has dramatically escalated its push toward artificial general intelligence (AGI), launching a new AI Superintelligence Lab and aggressively hiring top talent from rival firms, most notably OpenAI. This hiring spree includes at least eight high-profile researchers who worked on OpenAI’s o3 and GPT-4o teams. The move comes with eye-popping compensation offers: reports suggest Meta dangled multi-year packages worth up to $300 million, although Meta has denied the $300 million figure specifically. Still, reports suggest that offers in the range of $100 million in equity are becoming the new norm in high-stakes AI recruitment.

This recruitment drive is part of a broader reorganisation under Meta’s newly created Superintelligence Labs (MSL), combining existing AI divisions—including FAIR and the LLaMA development team—into one streamlined unit aimed squarely at AGI.

OpenAI, for its part, appears to be reeling. The high-profile departures have sparked internal concerns about the organisation’s culture and long-term vision. OpenAI execs likened Meta’s tactics to ‘breaking into our home,’ while others within OpenAI point to growing discontent over its shift to a capped-profit structure and Microsoft’s increasing influence. To contain the fallout, OpenAI is now reportedly revising its compensation and equity structures to retain remaining staff.

At the same time, tensions between OpenAI and Microsoft—its largest investor and infrastructure partner—are also coming to a head. A central point of contention is a clause in their 2019 partnership agreement that terminates Microsoft’s access to OpenAI’s technology if OpenAI achieves AGI. Microsoft is pushing to remove this clause as part of a renegotiation tied to OpenAI’s shift toward a for-profit model. Microsoft reportedly wants a more substantial equity stake—potentially up to 35%—and clearer influence over strategic direction.

These developments highlight an AI ecosystem undergoing deep transformation. Meta is betting on vertical integration, elite hiring, and open-source models as the path to AI dominance. OpenAI is grappling with internal coherence, external partnerships, and its public mission. Microsoft, meanwhile, is recalibrating its relationship with OpenAI even as competitors like Google DeepMind and Anthropic continue their own ambitious pushes.

Looking ahead, the implications are far-reaching. The extraordinary compensation packages now on offer are reshaping global talent flows, drawing researchers out of academia and away from startups toward Big Tech’s AGI arms race. As Meta centralises its superintelligence efforts and OpenAI charts its future course, governance questions loom large: Who decides when AGI is achieved? Who controls access? In the coming months, expect not only fierce competition for technical breakthroughs but also intensifying negotiations over power, accountability, and the rules of the game in an increasingly AGI-centric tech world.

Last week in Geneva

The Global Digital Collaboration Conference, held on July 1–2, 2025, in Geneva, Switzerland, was a significant event focused on advancing digital identity, credentials, and trusted infrastructure. Hosted by Federal Councillor Beat Jans and co-organised with 46 organisations, including UN agencies, international bodies, standardisation organisations, and open-source groups, the conference attracted over 1,000 experts from around the world. Discussions over 22 overview sessions and almost 100 collaborative sessions highlighted the importance of aligning governance frameworks with technical design to build trustworthy, open digital systems.

Diplo, the organisation that operates the Digital Watch Observatory, C4DT-EPFL and the Swiss Federal Department of Foreign Affairs (FDFA) co-hosted the session ‘Understanding cyber norms and the rules-based order: Ahat are the stakeholders’ roles in protecting critical infrastructure?” during the conference.

The UN CSTD multi-stakeholder working group on data governance at all levels kicked off its second meeting yesterday, 3 July, in Geneva. Yesterday, the group started discussing (a) fundamental principles of data governance at all levels as relevant for development; (b) proposals to support interoperability between national, regional and international data systems; (c) considerations of sharing the benefits of data; (d) options to facilitate safe, secure and trusted data flows, including cross-border data flows as relevant for development. They continued these discussions today and started discussions on the structure of the group’s report. The modalities of the group’s work, as well as agreement on dates and the provisional agenda for future sessions, are on the agenda for this afternoon.

DW team

For the main updates, reflections and events, consult the RADAR, the READING CORNER and the UPCOMING EVENTS section below.

Join us as we connect the dots, from daily updates to main weekly developments, to bring you a clear, engaging monthly snapshot of worldwide digital trends.

RADAR

Highlights from the week of 27 June – 4 July 2025

China unveils server chips to rival Intel models

Beijing narrows tech gap with CPU and quantum launches.

the flag of switzerland flying on a banner

Ransomware attack hits Swiss government data

Sarcoma hackers leaked 1.3TB of sensitive files after breaching Swiss contractor Radix.

africa ai digital continent only the african continent with artificial intelligence digitalised

Africa risks being left behind in global AI development

Oxford researchers reveal only 32 countries have the infrastructure to build advanced AI, leaving most of Africa sidelined in the race.

computer keyboard with red deepfake button key deepfake dangers online

Denmark proposes landmark law to protect citizens from deepfake misuse

Platforms could face fines for failing to remove illegal deepfake content under proposed Danish law.

Ransomware victims still paying, Sophos finds

Sophos says 49% of ransomware victims paid in 2025, but average ransom payments and backup use have declined.

UK plans new laws to tackle undersea cable sabotage

As geopolitical tensions mount beneath the waves, the UK is racing to future-proof its defence laws against unseen threats lurking in the deep.

Hacktivist attacks surge in Iran–Israel tensions

Experts say geopolitical hacktivism now poses serious risks to national infrastructure, calling for coordinated strategic cyber defences.

DeepSeek gains business traction despite security risks

US and global firms adopt DeepSeek’s models due to cost savings, even as public sector bans remain in place.

View all updates

UPCOMING EVENTS

BRICS 2025 Summit

The BRICS partnership will use the annual summit, to be held 6-7 July in Rio de Janeiro, Brazil, to address issues such as the environment, energy, science and technology, health and the inclusion of more civil society actors.

UN OEWG 2021-2025 11th substantive session

The Open-Ended Working Group (OEWG) on ICT security will hold its eleventh substantive session on 7-11 July 2025 in New York, USA. This will be the final session of the group’s work.

WSIS+20 High-Level Event 2025

WSIS+20 High-Level Event 2025 will take place on 7–11 July 2025 in Geneva, Switzerland. The event will facilitate multistakeholder dialogue on achievements, key trends, and challenges since the two phases of WSIS in 2003 and 2005. In the lead-up to the WSIS+20 review by the UN General Assembly, the event will also feature discussions on progress made in the implementation of the WSIS outcomes.

AI for Good Global Summit 2025

The AI for Good Global Summit 2025 will be held from 8 to 11 July in Geneva, Switzerland and feature 3 events: AI for Good Global Summit from 8 to 9 July, AI Governance Day on 10 July, and International AI Standards Day on 11 July.

Navigating the interplay between artificial intelligence, philosophy, education, and governance

This discussion between young leaders and experts will explore how AI can be governed in a way that reflects shared human values and ensures inclusive, sustainable, and ethical development.

Big week ahead—don’t miss a beat!
With WSIS+20 High-Level Event 2025, AI for Good 2025, and the final session of the OEWG all taking place next week, it’s impossible to follow everything live. That’s where we come in.

Our coverage combines expert analysis with AI-generated session reports and insights to help you stay informed, even when you can’t attend it all.

Bookmark our event pages to get real-time updates, highlights, and key takeaways as they happen.

READING CORNER

AI and Magical Realism: When technology blurs the line between wonder and reality – Diplo

The interplay between the magical and the real is at the heart of AI governance. How do we regulate something that feels both wondrous and mundane? How do we balance its promises with its perils? To navigate this, we might turn to the literary tradition of magical realism, where the impossible coexists seamlessly with the ordinary.

The future of global security and why cyber diplomacy matters – Diplo

At June’s G7 meeting, leaders agreed that reactive defense is no longer enough for AI and cybersecurity challenges. Clear rules to prevent conflict, but questions remain about who will set them and the Global South’s role in this new technological era.

Internet Governance Forum 2025

Revisit the key discussions of IGF 2025 with expert reporting with AI-powered tools, including session reports, a visual map of discussions, and an assistant that answers your policy-related questions.

ai generated image of a humanoid robot touching a digital hologram in a hyperreal urban background

The cognitive cost of AI: Balancing assistance and awareness

What if the very tool designed to boost your productivity is quietly dulling your mind each time you use it?

The power of self-organisation – Diplo

Can Wikipedia teach diplomacy a lesson? Aldo Matteucci contrasts rigid hierarchies with messy, adaptive self-organising systems, and asks which one truly gets more done.

IGF 2025 Highlights by DiploAI and experts

Dear readers,

This year, we reimagined how our IGF coverage can serve you.

At IGF 2025, we launched personalised reporting: participants told us which topics mattered most, and we delivered tailored, expert-verified updates—powered by DiploAI. It’s just one way we’ve deepened our commitment to timely, accessible, and actionable reporting.

Combined with interactive tools like a visual summary and a knowledge graph of the event, as well as a conversational AI assistant, our IGF 2025 coverage made navigating digital policy easier and more relevant than ever.
In partnership with the IGF Secretariat and the Government of Norway (as the host country), we were proud to provide AI-enabled, just-in-time reporting that brought the forum’s discussions directly to you.

Our session summaries deliver in-depth AI-powered insights for every discussion. Each report breaks down the key points, highlights the speakers’ perspectives, and maps out consensus and disagreements. Complementing this, our knowledge graphs visually chart the relationships among topics and contributors. To encourage ongoing dialogue, we also present thoughtful follow-up questions based on the session’s content.

With the support of DiploAI, our team provided daily summaries that distilled the essential insights from each day’s discussions.

Our AI system processed and analysed all session transcripts from the day, delivering concise overviews of key issues, the most important questions under discussion, and additional inputs that captured the daily highlights at the IGF.

Explore the event through our interactive visual overview, which maps the connections between major themes and participants. This engaging tool helps you see how different topics and conversations intersect across the forum. Whether you want to track particular speakers, explore your areas of interest, or spot broader patterns, the visual overview offers an intuitive and compelling way to uncover key insights from the forum.

Our AI assistant – powered by DiploAI – is designed to answer questions about IGF 2025 discussions. It can provide information about event speakers, arguments, transcripts, and other related queries.

We asked the tough questions—now it’s AI’s turn to respond. Drawing from key discussions, these answers won’t close the debate, but they’ll move it forward. They offer insights, spark reflection, and open the door to new ideas in digital policy.

Visit the IGF 2025 web page

Geneva Dialogue Community Newsletter #2

Dear friends,

Welcome to the second edition of the Geneva Dialogue Community Newsletter – this edition highlights three key trends shaping today’s cyber landscape.

First, there is a growing focus on cyber defence and cyberwarfare. The UK has announced the creation of a new Cyber and Electromagnetic Command, the USA is integrating generative AI into its cyber operations, and the Dutch Ministry of Defence is expanding its team of cyber reservists—demonstrating how digital capabilities are becoming central to national defence strategies.

Second, countries are taking important steps in setting new cybersecurity policy directions. Japan is developing a new national cybersecurity strategy, India has introduced AI incident reporting for critical infrastructure, and Denmark is working to reduce reliance on foreign software as part of its digital sovereignty efforts. Meanwhile, the EU and Australia are launching negotiations on a Security and Defence Partnership.

Finally, in the area of cyber norms and thanks to your contributions, the Geneva Dialogue has launched the second chapter of the Geneva Manual, focusing on how non-state stakeholders can implement the agreed cyber norms and confidence-building measures (CBMs) for the protection of critical infrastructure. This comes at a timely moment, as the UN’s Open-Ended Working Group (OEWG) – the primary forum for cyber discussions within the UN – approaches its concluding session in July.

As highlighted in Diplo’s analysis of the OEWG’s legacy (below), while progress has been made in reaffirming key norms and advancing voluntary cooperation, critical gaps remain, particularly in operationalising these norms and ensuring broader stakeholder engagement. The Geneva Dialogue’s work directly addresses this challenge by translating norms into actionable guidance and by reinforcing the role of industry, academia,the technical community, and civil society in supporting responsible state behaviour in cyberspace.

The Geneva Dialogue team

Highlights from the previous months

dig.watch

Breaking down the OEWG’s legacy: Hits, misses, and unfinished business

The OEWG on cybersecurity (2019–2025) shaped global debates on digital security—but did it deliver? External experts weigh in on its lasting impact, while our team, who tracked the process from day one, dissect the milestones and missed opportunities. Together, these perspectives reveal what’s next for cyber governance in a fractured world.

flags european union australia linen flags closeup flag made canvas eu australian canberra national symbols 3d illustration

dig.watch

EU and Australia to begin negotiations on security and defence partnership

Brussels and Australia will launch negotiations on a non-binding Security and Defence Partnership to enhance cooperation in areas such as cyber threats, counter-terrorism, and defence industry collaboration.

dig.watch

Supply chain cyber attack hits UBS and Swiss banks

Attackers published stolen internal data on dark web, UBS confirms.

dig.watch

Denmark moves to replace Microsoft software as part of digital sovereignty strategy

Denmark’s Ministry of Digitalisation has begun phasing out Microsoft Office 365 and Windows, replacing them with LibreOffice and Linux as part of a government-wide initiative to enhance digital sovereignty and reduce dependence on US technology providers.

dig.watch

UK National Cyber Security Centre calls for strategic cybersecurity policy agenda

The UK’s National Cyber Security Centre has urged the government to adopt a strategic cybersecurity policy agenda, citing legislative delays and insufficient market incentives for secure technology development.

dig.watch

India unveils AI incident reporting guidelines for critical infrastructure

As India positions itself as a global leader in AI safety, a new framework could quietly reshape how nations prepare for algorithmic failures in critical systems.

Geneva Dialogue at the African School of Internet Governance 1200x800 1

dig.watch

Diplo highlights cybersecurity at AfriSIG

As Africa’s digital landscape rapidly evolves, new alliances and fresh perspectives are emerging to shape the continent’s cybersecurity future from within.

dig.watch

Trump Executive Order revises US cyber policy and sanctions scope

President Donald Trump signed an Executive Order revising US cybersecurity policy by narrowing sanctions to foreign actors, revoking digital ID provisions, and directing new technical standards on encryption, AI, and software security.

ChatGPT Image May 29 2025 at 06 45 43 PM

dig.watch

UK to establish cyber and electromagnetic command to enhance warfare capabilities

As the digital battlefield expands, the UK is reshaping its military to confront invisible threats with a force designed for the age of cyber and electromagnetic warfare.

japan flag is depicted on a sports cloth fabric with many folds sport team waving banner

dig.watch

Japan to develop new cybersecurity strategy and measures

As Japan ramps up its cybersecurity efforts, a sweeping transformation is underway—one that could redefine how the country protects its digital infrastructure in the face of quantum-era threats and geopolitical tensions.

ChatGPT Image May 29 2025 at 06 24 15 PM

dig.watch

China blamed of cyberattack on Czech Republic government networks

The Czech Republic blames China-linked hacking group for a breach on its foreign ministry network. China has dismissed allegations but NATO, the EU and the US stand behind Prague.

netherlands flag is depicted on a sports cloth fabric with many folds sport team waving banner

dig.watch

Dutch Ministry of Defence expands recruitment of cyber reservists to support national cybersecurity efforts

The Dutch Ministry of Defense is expanding its recruitment of cyber reservists—civilian cybersecurity professionals who support military cyber operations on a part-time basis—as part of its strategy to enhance national cyber defense capabilities.

Geneva Dialogue Announcing Chapter II GM 2025

genevadialogue.ch

From words to protection – how do cyber norms guide security of critical infrastructure? Announcing Chapter II of the Geneva Manual

The session titled ‘From words to protection – how do cyber norms guide security of critical infrastructure?’ took place on 14 May 2025 in Geneva, as part of the Global Conference on Cyber Capacity Building (GC3B) organised by the Global Forum on Cyber Expertise (GFCE). This session marked the launch of Chapter II of the Geneva Manual – a milestone in defining what the UN cyber norms mean for real-world protection of critical infrastructure.

flag of the united states of america detail

dig.watch

US Cyber Command integrates generative AI for enhanced cybersecurity operations

The agency has outlined an AI roadmap with around 100 tasks to embed AI across logistics, security, and defence functions.

View all updates

DW Weekly #218 – DiploAI at the IGF 2025 in Norway, bridging the digital divide, AI, data governance, digital justice, and more!

20 – 27 June 2025

Book, Comics, Publication, Person, Adult, Male, Man, Clothing, Coat, Face, Head, Art

Dear readers,

This week, DiploAI took center stage in Lillestrøm, Norway, at the 20th Internet Governance Forum (IGF 2025), the world’s annual gathering on internet governance, where we proudly served as the official reporting team in partnership with the Government of Norway and the IGF Secretariat, combining Diplo’s diplomatic expertise with AI.

Armed with cutting-edge AI-driven transcription, our team captured every session, from the opening keynote to the final workshop, delivering real-time news, detailed reports, and daily newsletters that you can find on our dig.watch platform. With Dr Jovan Kurbalija and Sorina Teleanu leading the charge, Diplo documented and actively shaped the dialogue, advocating for inclusive, innovative governance alongside global leaders, policymakers, and tech visionaries.

DiploAI at the IGF 2025

The forum kicked off on 23 June with a bold focus on ‘Building Digital Governance Together,’ setting the stage for intense debates on AI’s role in humanity’s future. Day 1 spotlighted the urgent need to bridge the digital divide, with 2.6 billion people still offline, and explored sustainable data centre solutions to curb the ICT sector’s 2-4% contribution to global greenhouse gas emissions. Dr Jovan Kurbalija’s Day 0 address at the WGIG+20 session called for a renewed IGF mandate, while discussions pushed for permanent leadership and innovative funding to strengthen the forum ahead of the WSIS+20 review.

On Day 2, held on 24 June, showcased success stories like India’s Aadhaar system (80 million daily users) and Brazil’s PIX (saving $5.7 billion annually), though funding challenges loomed large. Dr Kurbalija spoke at an open forum on international AI cooperation, and Sorina Teleanu moderated a parliamentary session on digital policy, reinforcing multistakeholder collaboration. The narrative shifted toward power concentration and accountability, with sessions on AI therapists and digital emblems for conflict zones sparking fresh ideas.

As IGF 2025 progressed into Days 3 and 4, the focus sharpened on actionable outcomes. Discussions on AI governance emphasised local ecosystems and cultural diversity, while the knowledge ecology project, unlocking 19 years of IGF data with AI, promised deeper insights for the SDGs.

Efforts to tackle internet fragmentation definitely caught the attention of the audience on the last day of IGF 2025, addressing its technical, governance, and user experience dimensions. The session focused on implementing Article 29C of the Global Digital Compact (GDC) to prevent fragmentation, with stakeholders advocating for measurable frameworks and sustained dialogue via national and regional IGFs. Key concerns included the tension between seamless infrastructure and fractured user realities, raised by Gbenga Sesan. Marilia Maciel highlighted the need for economic research to understand fragmentation’s real-world impact, noting that billions face access and content restrictions, underscoring the issue as both a technical and human challenge requiring collaborative action.

Freedom Online Coalition convened a diverse panel to explore ‘How can technical standards bridge or broaden the digital divide?‘ The session, moderated by Laura O’Brien, Senior International Counsel at Access Now, highlighted how open and interoperable standards can empower underserved communities.

The final day reframed the debate with critical questions: What should the IGF look like post-2025, and how can it align with the Global Digital Compact? With Diplo’s booth (#45) and the CADE consortium’s (#57) presence in the IGF Village until the end of today’s sessions, we invite you to explore these instructive expert dialogues on dig.watch in the special IGF 2025 event section for more detailed information!

EuroDIG outcomes shared at IGF 2025 session in Norway

UNESCO and ICANN lead push for multilingual and inclusive internet governance

Ask our personalised IGF 2025 chatbot for more information on which were the most important topics discussed at the forum!

Other highlights of the week:

OpenAI chief executive Sam Altman revealed that he had a conversation with Microsoft CEO Satya Nadella on Monday to discuss the future of their partnership.

A groundbreaking quantum leap has taken place in space exploration. The world’s first photonic quantum computer has successfully entered orbit aboard SpaceX’s Transporter 14 mission.

Bitcoin holds strong above $100,000. Technical signals on Bitcoin and Ethereum point to cautious optimism with key resistance levels ahead.

The digital sector is urging the EU leaders to delay the AI act, citing missing guidance and legal uncertainty.

Meta Platforms’ messaging service WhatsApp has been banned from all devices used by the US House of Representatives, according to an internal memo distributed to staff on Monday.

NATO’s 76th summit opened in the Hague amid rising tensions in EU and the Middle East, overshadowed by conflict and cyber threats.

The National Cyber Security Centre (NCSC) has published new guidance to assist organisations in meeting the upcoming EU Network and Information Security Directive (NIS2) requirements.

A coalition of cybersecurity agencies, including the NSA, FBI, and CISA, has issued joint guidance to help organisations protect AI systems from emerging data security threats.

Diplo Blog – AI in Sophie’s world: How a philosophy book can help us govern AI

Jovan Kurbalija

As we convene in Oslo for the Internet Governance Forum, we reflect on the philosophical insights from Jostein Gaarder’s ‘Sophie’s World.’ The novel’s exploration of identity and constructed reality parallels the challenges of AI governance today. Just as Sophie learns her existence is scripted, AI systems are shaped by data and algorithms, raising questions about consciousness and accountability. The complexities of AI, such as the ‘black box’ problem, highlight the need for transparency and understanding. Ultimately, governance should prioritise values and meaning, urging us to engage with fundamental questions about identity and existence in our digital age. Read the full blog!

Last week in Geneva

HRC

The ongoing 59th session of the Human Rights Council (HRC), which opened on 16 June and will last until 11 July 2025 at the Palais des Nations in Geneva, presided over by Ambassador Jürg Lauber of Switzerland, addressed critical issues including civil society space, climate change, sexual orientation and gender identity, and violence against women, while spotlighting grave human rights situations in countries like Afghanistan, China, and Sudan.

Key outcomes included the consideration of OHCHR reports on 26 June, such as the practical application of the Guiding Principles on Business and Human Rights to technology companies’ AI activities (A/HRC/59/32) and the use of digital technologies for universal birth registration (A/HRC/59/61), signaling a focus on tech accountability and inclusion.

The session also featured interactive dialogues on systemic racism, the rights of Indigenous peoples, and climate change, with the Special Rapporteur Elisa Morgera presenting a report on defossilizing economies (A/HRC/59/42), though debates revealed tensions between economic priorities and human rights obligations. Despite robust discussions, scepticism persists about the practical impact of these deliberations amid geopolitical divides and funding challenges for the UN rights system.

ITU Council 2025

ITU Council 2025, which opened on 17 June and is closing today, 27 June, in Geneva, Switzerland, delivered key highlights and outcomes that are shaping the future of global telecommunications.

The session focused on critical governance and operational matters, including the review and adoption of the 2026–2027 biennial budget, oversight of ITU programs, and strategic planning for the 2028–2031 cycle, ensuring alignment with evolving telecom landscapes.

A significant milestone was the inauguration of the Giga Connectivity Centre on 25 June at Geneva’s Campus Biotech, a joint ITU-UNICEF initiative co-hosted with Switzerland and Spain, aimed at connecting every school to the internet.

The Council also confirmed the World Telecommunication Development Conference for 17–28 November 2025 in Baku and set the 202 6 World Telecommunication/ICT Policy Forum theme as ‘Accelerating an inclusive, sustainable, resilient, and innovative digital future,’ emphasising bridging digital divides.

Additionally, an AI Workshop for Councillors on 16 June explored opportunities in diplomacy and governance, while side events like the WSIS information session and AI for Good initiative underscored ITU’s commitment to digital transformation and sustainable development goals.

Upcoming events

For the main updates, reflections and events, consult the RADAR, the READING CORNER and the UPCOMING EVENTS section below.

Join us as we connect the dots, from daily updates to main weekly developments, to bring you a clear, engaging monthly snapshot of worldwide digital trends.

DW Team

RADAR

Highlights from the week of 20 – 27 June 2025

IGF 2025 opens in Norway with focus on inclusive digital governance

Internet Governance Forum returns to Europe for its 20th anniversary.

Big Tech’s grip on information sparks urgent debate at IGF 2025 in Norway

With democracy under siege from AI-fueled disinformation and opaque algorithms, global leaders at IGF 2025 grappled with the urgent question: who really controls the truth in our digital age?

Civil society pushes for digital rights and justice in WSIS+20 review at IGF 2025

As global civil society leaders converge to reshape the digital rights landscape, their bold proposals signal a pivotal moment in the future of internet governance.

Children safety online in 2025: Global leaders demand stronger rules

IGF panel warns that regulation without enforcement won’t protect children from digital exploitation.

WSIS prepares for Geneva as momentum builds for impactful digital governance

At IGF 2025 in Norway, stakeholders mapped priorities ahead of the WSIS+20 high-level event in Geneva.

WGIG reunion sparks calls for reform at IGF 2025 in Norway

As digital power dynamics shift and global tensions rise, a reunion of internet governance pioneers reignites debate over who really shapes the rules of our online future.

Kurbalija’s book on internet governance turns 20 with new life at IGF

Amid rapid digital change and rising AI hype, a familiar voice returns to challenge the buzzwords and bring clarity to the evolving debate.

Global consensus grows on inclusive and cooperative AI governance at IGF 2025

AI is no longer confined to tech labs—it’s becoming a global concern that demands cooperation, creativity, and a shared commitment to humanity’s future.

AI governance debated at IGF 2025: Global cooperation meets local needs

AI governance must be inclusive, context-aware, and rooted in human rights, IGF 2025 panellists agree.

AI and the future of work: Global forum highlights risks, promise, and urgent choices

As AI continues to reshape industries and redefine the workplace, world leaders and innovators face a pressing question: Can we ensure its benefits are shared fairly before the gap between…

IGF panel urges rethinking internet governance amid rising geopolitical tensions

Can imagining the internet of 2040 help us fix the broken governance models of today?

AU Open Forum at IGF 2025 highlights urgent need for action on Africa’s digital future

With Africa’s digital ambitions at a crossroads, leaders at the IGF 2025 urged a decisive shift from promises to practical solutions that can bridge the continent’s growing connectivity and policy…

Tower of Babel reimagined: IGF 2025 experiment highlights language barriers in internet governance

What happens when global internet policy experts suddenly abandon English and try to govern in dozens of different native languages?

Rethinking AI in journalism with global cooperation

As AI rapidly transforms journalism, a diverse group of global experts and grassroots voices came together in Norway to rethink how media can harness these tools without compromising ethics, inclusion,…

View all updates

READING CORNER