The District Court of The Hague has rejected an attempt by three Dutch citizens to block the government from renewing its contract with Solvinity, the company responsible for hosting and technically managing systems linked to DigiD.
The plaintiffs argued that Solvinity’s planned acquisition by US-based IT provider Kyndryl could place sensitive data from more than 16 million DigiD users under US jurisdiction, potentially exposing it to US authorities and creating risks to critical public services such as healthcare, pensions, taxes, and unemployment systems.
Despite these concerns, the court ruled in favour of the Dutch State, allowing the agreement to proceed. Judges did not accept arguments that the deal would immediately threaten data security or justify halting the contract.
The decision leaves further scrutiny to the Investment Assessment Office, which is reviewing national security risks linked to the acquisition. The case highlights ongoing tensions around digital sovereignty and data protection in the Netherlands.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
Privacy rights group noyb has filed a complaint against LinkedIn, alleging that the platform restricts access to certain user data by placing it behind a paid Premium subscription.
The complaint centres on LinkedIn’s ‘Who’s viewed your profile’ feature, which shows users who have visited their profile. According to noyb, LinkedIn tracks profile visits and makes detailed visitor information available to Premium subscribers, while refusing to provide the same data free of charge when users submit an access request under Article 15 of the GDPR.
Noyb argues that users have the right to receive their own personal data free of charge under the EU data protection rules. The organisation claims that LinkedIn has cited data protection concerns when refusing access requests, despite making similar information available through its paid subscription service.
The complaint was lodged with the Austrian Data Protection Authority and seeks enforcement action requiring LinkedIn to provide the data requested, as well as potential penalties. Noyb also questions whether LinkedIn’s tracking of profile visits complies with the EU consent requirements.
LinkedIn has reportedly denied the allegations, saying it complies with applicable rules and provides relevant information in accordance with its privacy policies.
The case adds to ongoing scrutiny of how digital platforms handle data access rights in the EU, particularly when information collected about users is also used for paid services.
Why does it matter?
The complaint tests whether platforms can monetise access to information that may also fall under users’ GDPR right of access. If regulators side with noyb, the case could affect how subscription-based platforms structure premium features that involve personal data, especially when the same data is withheld from non-paying users who make formal access requests.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
Apple is reportedly preparing a major expansion of Apple Intelligence that could allow users to choose which AI model powers Siri and other system features. According to recent reports, iOS 27, iPadOS 27, and macOS 27 may introduce a new ‘Extensions’ framework designed to integrate third-party AI systems directly into Apple’s software ecosystem.
The reported feature would allow applications such as Gemini and Claude to connect with Siri through their App Store apps. Users may be able to select different AI providers for different tasks, while Apple is also said to be testing separate Siri voices for responses generated by external models rather than Apple’s own systems.
The move would expand Apple’s broader AI partnership strategy rather than replace existing integrations. ChatGPT already supports selected Apple Intelligence functions, and earlier reporting suggested Google Gemini could eventually power parts of Siri itself. The new framework appears aimed at turning Apple devices into a wider AI platform that supports multiple large language models rather than a single assistant stack.
Apple is expected to present further details during its Worldwide Developers Conference on 8 June 2026. If the reported changes materialise, they could significantly reshape how users interact with AI assistants by giving them more control over which models handle tasks such as search, writing, and image generation.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
Apple has agreed to pay $250 million to settle a class action lawsuit alleging that it misled consumers about the readiness and availability of AI-powered Siri features promoted ahead of the iPhone 16 launch. Under the proposed agreement, eligible US customers who bought supported iPhone models between 10 June 2024 and 29 March 2025 may receive between $25 and $95 per device, depending on the number of claims. Apple denied wrongdoing and settled the case without admitting liability.
The complaint argued that consumers who purchased supported iPhone 15 and iPhone 16 models expected advanced Apple Intelligence features and a significantly upgraded Siri experience that were not available at the time of sale. Plaintiffs said Apple’s marketing created the impression that the new capabilities would arrive sooner and with broader functionality than users ultimately received.
The settlement comes shortly before Apple’s annual Worldwide Developers Conference, where the company is widely expected to present further updates to Siri and its wider AI strategy.
Why does it matter?
The case shows how AI product marketing is becoming a legal and regulatory risk, not just a branding issue. As technology companies use generative AI features to drive device sales and platform adoption, courts and consumers are paying closer attention to whether those capabilities are actually available when products reach the market. The Apple settlement suggests that overstating AI readiness can create liability even before regulators step in, making transparency around launch claims increasingly important across the sector.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
DeepSeek has again placed itself at the centre of the global AI race. After drawing worldwide attention with its R1 reasoning model in early 2025, the Chinese company has recently released DeepSeek V4, a new model designed to compete not only on performance, but also on price, openness and efficiency.
The hype around DeepSeek V4 is not based on a single feature. The model comes with a 1 million-token context window, open weights, two versions for different use cases and a strong focus on agentic workflows such as coding, research, document analysis and long-running tasks. In a market still dominated by expensive closed models, DeepSeek is trying to prove that powerful AI does not need to remain locked behind trademarked systems.
A model built for long memory
The most immediate difference between DeepSeek V4 and other models is context length. Both DeepSeek-V4-Pro and DeepSeek-V4-Flash support a 1-million-token context window, meaning they can process inputs far longer than those of older generations of mainstream models. According to DeepSeek’s official release, one million tokens is now the default across all official DeepSeek services.
For ordinary users, that may sound technical. In practice, it matters because a longer context allows models to work with large documents, long conversations, full codebases, legal materials, research archives or complex project histories without losing track as quickly.
That is why DeepSeek V4 is not just another chatbot release. It is aimed at the next stage of AI use, where models are expected to act less like question-answering tools and more like assistants that can follow long processes over time.
Two models for two different needs
DeepSeek V4 comes in two main versions. DeepSeek-V4-Pro is a larger and more capable model, with 1.6 trillion total parameters and 49 billion active parameters. DeepSeek-V4-Flash is a smaller model, with 284 billion total parameters and 13 billion active parameters, designed for faster and more cost-effective workloads.
That distinction is important. Not every user needs the strongest model for every task. A company summarising documents, routing queries or running basic support may choose Flash. A developer working on complex coding tasks, long-context agents or advanced reasoning may prefer Pro.
DeepSeek’s release reflects a broader trend in AI. The best model is no longer always the biggest one. Cost, speed, context size and deployment flexibility are now as important as raw benchmark performance.
Why the price matters
One reason DeepSeek attracts so much attention is its aggressive pricing. DeepSeek’s API page lists V4-Flash at USD 0.14 per 1 million input tokens on a cache miss and USD 0.28 per 1 million output tokens. V4-Pro is listed at USD 1.74 per 1 million input tokens and USD 3.48 per 1 million output tokens before the temporary 75% discount.
For developers and companies, that changes the calculation. High-performing AI models are useful only if they can be deployed at scale. If every long document, coding session or agentic workflow becomes too expensive, adoption slows down.
DeepSeek’s challenge to the market is therefore not only technical. It is economic. The company is pushing the idea that frontier-level AI should be cheaper to run, easier to access and less dependent on closed ecosystems.
The architecture behind the hype
DeepSeek V4 uses a mixture-of-experts approach, meaning only part of the model is active during each response. That helps explain why the model can be very large on paper, yet still more efficient to run than a dense model of similar overall size.
The more interesting part is how DeepSeek handles long context. NVIDIA’s technical overview explains that DeepSeek V4 uses hybrid attention, combining compression and selective attention techniques to reduce the cost of processing very long prompts. NVIDIA says these changes are designed to cut per-token inference FLOPs by 73% and reduce KV cache memory burden by 90% compared with DeepSeek-V3.2.
For a non-technical audience, the point is simple. DeepSeek V4 is trying to solve one of the biggest problems in modern AI: how to make models remember and process much more information without becoming too slow or too expensive.
That is where much of the hype comes from. The model is not merely larger. It is designed around the economics of long-context AI.
Why NVIDIA is still in the picture
NVIDIA’s role in the DeepSeek V4 story is especially interesting. DeepSeek is often discussed as part of China’s effort to build a more independent AI ecosystem, but NVIDIA has also been quick to move forward to support developers who want to build with the model.
In its technical blog, NVIDIA describes DeepSeek V4 as a model family designed for efficient inference of million-token contexts. The company says DeepSeek-V4-Pro and V4-Flash are available through NVIDIA GPU-accelerated endpoints, while developers can also use NVIDIA Blackwell, NIM containers, SGLang and vLLM deployment options.
NVIDIA also reports that early tests of DeepSeek-V4-Pro on the GB200 NVL72 platform showed more than 150 tokens per second per user. That matters because long-context models place heavy memory pressure, as well as on compute and networking infrastructure. The model may be efficient by design, but serving it at scale still requires serious hardware.
So, DeepSeek V4 does not remove NVIDIA from the story – it complicates it. The model is part of a broader push towards more efficient AI, but the infrastructure race remains central.
The chip question behind the model
DeepSeek V4 also arrives at a time when AI infrastructure is becoming just as important as model performance. MIT Technology Review frames the release partly through that lens, noting that DeepSeek’s new model reflects China’s broader attempt to reduce reliance on foreign AI hardware and build a more self-sufficient technology stack.
That detail matters because the AI race is no longer only about who builds the most capable model. It is also about who controls the chips, software frameworks and data centres needed to run it.
Replacing NVIDIA, however, remains difficult. Its advantage lies not just in its chips, but also in the software ecosystem developers have built around its platforms over many years. Moving to alternative hardware means adapting code, rebuilding tools and proving that the new systems are stable enough for serious use.
DeepSeek V4, however, sits between two realities. It points towards China’s ambition to build a more independent AI stack, while NVIDIA’s rapid support for the model shows that frontier AI still depends heavily on established infrastructure.
Open weights as a strategic move
DeepSeek V4 is also important because the model weights are available through Hugging Face under the MIT License. That gives developers more freedom to inspect, adapt and deploy the model than they would have with a fully closed commercial system.
Open-weight models are becoming a major pressure point in the AI race. Closed models may still lead in some areas, especially in polished consumer products, enterprise support and safety layers. However, open models offer something different: flexibility.
For universities, start-ups, smaller companies and developers outside the largest AI ecosystems, that flexibility matters. It means advanced AI can be tested, modified and integrated without relying entirely on a handful of dominant providers.
Benchmarks need caution
DeepSeek presents V4-Pro as highly competitive across reasoning, coding, long-context and agentic benchmarks. Hugging Face lists results including 80.6 on SWE-bench Verified, 90.1 on GPQA Diamond and 87.5 on MMLU-Pro for DeepSeek-V4-Pro.
Those numbers are impressive, but they should not be treated as the full story. Benchmarks are useful, but they rarely capture every real-world use case. A model can score well on coding tests and still struggle with reliability, factual accuracy, safety or complex multi-step workflows in production.
That caution is important. The AI industry often turns benchmarks into headlines, while real performance depends on deployment, prompting, safety controls and the specific task at hand.
More than just another model release
DeepSeek V4 matters because it combines several trends into one release: long context, lower prices, open weights, agentic workflows and geopolitical competition. It also shows that the AI race is no longer fought only in labs, benchmarks and data centres. Visibility now matters too. Tools such as Diplo’s Digital Footprints show how digital presence shapes the way technology actors and media narratives are discovered, ranked and understood. At this stage, the competition is not only about who has the smartest model. It is also about who can make intelligence cheaper, more available and easier to deploy.
That does not mean DeepSeek has solved every problem. Questions remain around independent benchmarking, safety, data governance, infrastructure and the broader political context of Chinese AI development. Still, the release does show where the market is heading.
The next phase of AI may not be defined solely by the most powerful model. It may be defined by the model that is powerful enough, affordable enough and open enough to change how people build products, services and tools with AI.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
Kazakhstan’s financial regulator has warned that several major cryptocurrency exchanges are operating without the licences required under the country’s current digital asset framework, reinforcing its strict authorisation regime.
The Astana Financial Services Authority identified prominent platforms, including HTX, Bitget, OKX, and MEXC, as operating without the necessary permits. Under existing rules, only entities licensed within the Astana International Financial Centre are allowed to provide regulated digital asset services.
Authorities stressed that international popularity does not exempt platforms from complying with local law. They also warned that unauthorised exchanges can expose users to financial losses, data breaches, and fraudulent schemes, and urged the public to verify platforms through the official register of licensed firms. AFSA’s website currently shows a regulated ecosystem with dozens of authorised entities across the AIFC framework.
The warning comes amid broader enforcement efforts as Kazakhstan tries to formalise its crypto sector while positioning itself as a regulated regional hub for digital assets. In parallel, law enforcement agencies have reported wider crackdowns on illegal crypto activity, including shadow exchanges and money-laundering networks.
Why does it matter?
Kazakhstan’s tightening enforcement shows a broader push to bring crypto activity into a more formal and supervised market structure. By restricting unlicensed platforms and steering users towards authorised entities, the authorities are trying to reduce exposure to financial crime, improve market transparency, and build credibility for Kazakhstan’s ambition to become a regulated regional digital asset hub.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
Cybercriminal activity tends to intensify during tax-return season, as taxpayers face tighter deadlines and share sensitive financial information. A recent Kaspersky analysis highlights the growing use of fake tax authority websites, phishing emails, and malicious downloads designed to steal personal and banking data.
Attackers are impersonating official revenue services across multiple countries, creating convincing portals that mimic government branding and online tax services. Victims are often prompted to enter login credentials, payment details, or download files containing malware aimed at compromising devices or extracting sensitive information.
Crypto holders are also being targeted through fake compliance portals and fraudulent regulatory notices. These schemes try to trick users into revealing wallet recovery phrases or linking digital wallets, which can lead to full asset theft once access is granted.
AI adds another layer of risk. Kaspersky warns that users who upload tax documents or personal financial data to unverified AI platforms may expose confidential information to leakage, misuse, or further fraud. More broadly, AI is also making phishing and impersonation campaigns easier to scale and harder to detect.
Security experts recommend relying only on official tax channels, checking websites and email sources carefully, avoiding unsolicited downloads, and using secure storage and trusted protection tools when handling tax documents.
Why does it matter?
Tax-season phishing campaigns show how financial data is increasingly being treated as a high-value target for cybercrime. As tax systems, digital finance, crypto assets, and AI tools overlap more closely, a single successful scam can lead not only to immediate financial loss but also to identity theft, device compromise, and broader damage to trust in digital services.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
China has introduced new online marketing rules for financial products, further tightening its long-standing restrictions on cryptocurrency-related activity. The new framework limits the promotion of financial products to licensed entities and treats digital currency trading and issuance as illegal financial activity.
Issued by the People’s Bank of China and seven other regulators, the Administrative Measures for Online Marketing of Financial Products will take effect on 30 September 2026. The rules extend responsibility to platforms, intermediaries, and content creators who promote or facilitate financial products online.
Any assistance in promoting or facilitating prohibited financial activity may now be treated as participation in illegal finance, expanding enforcement beyond direct trading bans. In practice, that broadens the focus from financial products themselves to the wider digital promotion layer, including online displays, traffic generation, and other forms of internet-based marketing support.
Authorities say the measures are intended to protect consumers by limiting misleading or aggressive online promotion, including livestream marketing and viral investment content. In that sense, the rules are not only about crypto, but about tighter control over how financial products are marketed in digital environments.
The policy also reinforces China’s existing position, dating back to 2021, when regulators declared all cryptocurrency transactions illegal, while pushing enforcement deeper into the digital advertising and distribution layers of financial markets.
Why does it matter?
Stronger oversight of online financial promotion shows that crypto-related advertising is increasingly being treated as a regulatory risk category, not just a marketing issue. The Chinese move also points to a broader trend in which regulators are extending scrutiny beyond financial products themselves to the digital channels, influencers, and platforms that help distribute them.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
The UK’s Financial Conduct Authority has led its first coordinated crackdown on illegal crypto trading, targeting firms operating without authorisation. The action forms part of wider efforts to enforce compliance in the sector.
According to the Authority, the operation involved identifying and taking action against companies that unlawfully promoted or offered crypto services. The move aims to protect consumers from potential risks.
The regulator stated that illegal crypto promotions can expose users to financial harm and undermine market trust. It emphasised the importance of ensuring firms meet regulatory requirements before operating.
The Authority said the crackdown reflects a stronger enforcement approach to unauthorised crypto activity, with further action expected to support market integrity in the UK.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
UNESCO has launched a new regional platform on AI in education for Latin America and the Caribbean, aiming to help governments respond to both a deep learning crisis and the rapid spread of AI tools in schools and universities.
Called the Observatory on Artificial Intelligence in Education for Latin America and the Caribbean, the initiative was launched on 14 April in Santiago, Chile, during the 2026 Forum of the Countries of Latin America and the Caribbean on Sustainable Development.
UNESCO presents the Observatory as the first regional platform anchored in the UN system dedicated to AI in education in Latin America and the Caribbean. It is designed as a multistakeholder mechanism bringing together the region’s 33 ministries of education, along with universities, research centres, teachers, and strategic partners, to generate evidence, strengthen capacities, and support public decision-making on how AI should be used in education.
The initiative is being framed as a response to two pressures at once. UNESCO says the region faces a serious learning crisis, while AI tools are spreading rapidly through classrooms and education systems, with uneven guidance and limited institutional preparedness. In that context, the Observatory is meant to support more context-specific policy development, stronger teacher training, and classroom-tested innovation within ethical frameworks, rather than leaving AI adoption to fragmented local experimentation.
That gives the launch a significance beyond a standard education technology initiative. The core argument is not simply that AI should be introduced into schools, but that governments need a shared regional capacity to shape its use. UNESCO sums that up with a simple principle: AI should not govern education; education should govern AI.
The Observatory is being developed with a broad coalition of regional and international partners, including the Development Bank of Latin America and the Caribbean, Chile’s National Centre for Artificial Intelligence, the Regional Centre for Studies on the Development of the Information Society, ECLAC, the Ceibal Foundation, Fundación Santillana, Tecnológico de Monterrey, ProFuturo, the Universidad del Desarrollo in Chile, and the International Research Centre on Artificial Intelligence. Its advisory council also includes the OECD, the Organisation of Ibero-American States, experts from Harvard University, and the UN Independent International Scientific Panel on AI.
Why does it matter?
The story shows UNESCO moving from broad principles on ethical AI to a more concrete regional governance model. Rather than issuing another general call for responsible AI in education, it is trying to build an institutional platform that can connect evidence, policy, teacher capacity, and public oversight across Latin America and the Caribbean.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
