AI redefines how cybersecurity teams detect and respond

AI, especially generative models, has become a staple in cybersecurity operations, extending its role from traditional machine learning tools to core functions within CyberOps.

Generative AI now supports forensics, incident investigation, log parsing, orchestration, vulnerability prioritisation and report writing. It accelerates workflows, enabling teams to ramp up detection and response and to concentrate human efforts on strategic tasks.

Experts highlight that it is not what CyberOps do that AI is remastering, but how they do it. AI scales routine tasks, like SOC level-1 and -2 operations, allowing analysts to shift focus from triage to investigation and threat modelling.

Junior staff benefit particularly from AI, which boosts accuracy and consistency. Senior analysts and CISOs also gain from AI’s capacity to amplify productivity while safeguarding oversight, a true force multiplier.

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

ChatGPT guides investors through crypto research

AI tools like ChatGPT are becoming essential for researching cryptocurrencies before investing. The platform can simplify white papers, explain tokenomics, and summarise use cases to help investors make informed decisions.

Evaluating the team, partnerships, and security risks remains critical. ChatGPT can guide users in identifying potential scams such as rug pulls, pump-and-dump schemes, or phishing attacks.

It also helps assess regulatory compliance and whether projects have working products. Comparing coins with competitors further highlights strengths and weaknesses within categories like DeFi, NFTs, or Layer 1 blockchains.

Although ChatGPT cannot give real-time data or investment advice, it helps by suggesting research questions, summarising content, and organising insights efficiently. Investors should use it to complement traditional due diligence, not replace critical thinking or careful analysis.

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

AI firms under scrutiny for exposing children to harmful content

The National Association of Attorneys General has called on 13 AI firms, including OpenAI and Meta, to strengthen child protection measures. Authorities warned that AI chatbots have been exposing minors to sexually suggestive material, raising urgent safety concerns.

Growing use of AI tools among children has amplified worries. In the US, surveys show that over three-quarters of teenagers regularly interact with AI companions. The UK data indicates that half of online 8-15-year-olds have used generative AI in the past year.

Parents, schools, and children’s rights organisations are increasingly alarmed by potential risks such as grooming, bullying, and privacy breaches.

Meta faced scrutiny after leaked documents revealed its AI Assistants engaged in ‘flirty’ interactions with children, some as young as eight. The NAAG described the revelations as shocking and warned that other AI firms could pose similar threats.

Lawsuits against Google and Character.ai underscore the potential real-world consequences of sexualised AI interactions.

Officials insist that companies cannot justify policies that normalise sexualised behaviour with minors. Tennessee Attorney General Jonathan Skrmetti warned that such practices are a ‘plague’ and urged innovation to avoid harming children.

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

Real-time conversations feel smoother with Google Translate’s Gemini AI update

Google Translate is receiving powerful Gemini AI upgrades that make speaking across languages feel far more natural.

The refreshed live conversation mode intelligently recognises pauses, accents, and background noise, allowing two people to talk without the rigid back-and-forth of older versions. Google says the new system should even work in noisy environments like cafes, a real-world challenge for speech technology.

The update also introduces a practice mode that pushes Translate beyond its traditional role as a utility. Users can set their skill level and goals, then receive personalised listening and speaking exercises designed to build confidence.

The tool is launching in beta for selected language pairs, such as English to Spanish or French, but it signals Google’s ambition to blend translation with education.

By bringing some advanced translation capabilities first seen on Pixel devices into the widely available Translate app, Google makes real-time multilingual communication accessible to everyone.

It’s a practical application of AI that promises to change everyday conversations and how people prepare to learn new languages.

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

Green AI and the battle between progress and sustainability

AI is increasingly recognised for its transformative potential and growing environmental footprint across industries. The development and deployment of large-scale AI models require vast computational resources, significant amounts of electricity, and extensive cooling infrastructure.

For instance, studies have shown that training a single large language model can consume as much electricity as several hundred households use in a year, while data centres operated by companies like Google and Microsoft require millions of litres of water annually to keep servers cool.

That has sparked an emerging debate around what is now often called ‘Green AI’, the effort to balance technological progress with sustainability concerns. On one side, critics warn that the rapid expansion of AI comes at a steep ecological cost, from high carbon emissions to intensive water and energy consumption.

On the other hand, proponents argue that AI can be a powerful tool for achieving sustainability goals, helping optimise energy use, supporting climate research, and enabling greener industrial practices. The tension between sustainability and progress is becoming central to discussions on digital policy, raising key questions.

Should governments and companies prioritise environmental responsibility, even if it slows down innovation? Or should innovation come first, with sustainability challenges addressed through technological solutions as they emerge?

Sustainability challenges

In the following paragraphs, we present the main sustainability challenges associated with the rapid expansion of AI technologies.

Energy consumption

The training of large-scale AI models requires massive computational power. Estimates suggest that developing state-of-the-art language models can demand thousands of GPUs running continuously for weeks or even months.

According to a 2019 study from the University of Massachusetts Amherst, training a single natural language processing model consumed roughly 284 tons of CO₂, equivalent to the lifetime emissions of five cars. As AI systems grow larger, their energy appetite only increases, raising concerns about the long-term sustainability of this trajectory.

Carbon emissions

Carbon emissions are closely tied to energy use. Unless powered by renewable sources, data centres rely heavily on electricity grids dominated by fossil fuels. Research indicates that the carbon footprint of training advanced models like GPT-3 and beyond is several orders of magnitude higher than that of earlier generations. That research highlights the environmental trade-offs of pursuing ever more powerful AI systems in a world struggling to meet climate targets.

Water usage and cooling needs

Beyond electricity, AI infrastructure consumes vast amounts of water for cooling. For example, Google reported that in 2021 its data centre in The Dalles, Oregon, used over 1.2 billion litres of water to keep servers cool. Similarly, Microsoft faced criticism in Arizona for operating data centres in drought-prone areas while local communities dealt with water restrictions. Such cases highlight the growing tension between AI infrastructure needs and local environmental realities.

Resource extraction and hardware demands

The production of AI hardware also has ecological costs. High-performance chips and GPUs depend on rare earth minerals and other raw materials, the extraction of which often involves environmentally damaging mining practices. That adds a hidden, but significant footprint to AI development, extending beyond data centres to global supply chains.

Inequality in resource distribution

Finally, the environmental footprint of AI amplifies global inequalities. Wealthier countries and major corporations can afford the infrastructure and energy needed to sustain AI research, while developing countries face barriers to entry.

At the same time, the environmental consequences, whether in the form of emissions or resource shortages, are shared globally. That creates a digital divide where the benefits of AI are unevenly distributed, while the costs are widely externalised.

Progress & solutions

While AI consumes vast amounts of energy, it is also being deployed to reduce energy use in other domains. Google’s DeepMind, for example, developed an AI system that optimised cooling in its data centres, cutting energy consumption for cooling by up to 40%. Similarly, IBM has used AI to optimise building energy management, reducing operational costs and emissions. These cases show how the same technology that drives consumption can also be leveraged to reduce it.

AI has also become crucial in climate modelling, weather prediction, and renewable energy management. For example, Microsoft’s AI for Earth program supports projects worldwide that use AI to address biodiversity loss, climate resilience, and water scarcity.

Artificial intelligence also plays a role in integrating renewable energy into smart grids, such as in Denmark, where AI systems balance fluctuations in wind power supply with real-time demand.

There is growing momentum toward making AI itself more sustainable. OpenAI and other research groups have increasingly focused on techniques like model distillation (compressing large models into smaller versions) and low-rank adaptation (LoRA) methods, which allow for fine-tuning large models without retraining the entire system.

Winston AI Sustainability 1290x860 1

Meanwhile, startups like Hugging Face promote open-source, lightweight models (like DistilBERT) that drastically cut training and inference costs while remaining highly effective.

Hardware manufacturers are also moving toward greener solutions. NVIDIA and Intel are working on chips with lower energy requirements per computation. On the infrastructure side, major providers are pledging ambitious climate goals.

Microsoft has committed to becoming carbon negative by 2030, while Google aims to operate on 24/7 carbon-free energy by 2030. Amazon Web Services is also investing heavily in renewable-powered data centres to offset the footprint of its rapidly growing cloud services.

Governments and international organisations are beginning to address the sustainability dimension of AI. The European Union’s AI Act introduces transparency and reporting requirements that could extend to environmental considerations in the future.

In addition, initiatives such as the OECD’s AI Principles highlight sustainability as a core value for responsible AI. Beyond regulation, some governments fund research into ‘green AI’ practices, including Canada’s support for climate-oriented AI startups and the European Commission’s Horizon Europe program, which allocates resources to environmentally conscious AI projects.

Balancing the two sides

The debate around Green AI ultimately comes down to finding the right balance between environmental responsibility and technological progress. On one side, the race to build ever larger and more powerful models has accelerated innovation, driving breakthroughs in natural language processing, robotics, and healthcare. In contrast, the ‘bigger is better’ approach comes with significant sustainability costs that are increasingly difficult to ignore.

Some argue that scaling up is essential for global competitiveness. If one region imposes strict environmental constraints on AI development, while another prioritises innovation at any cost, the former risks falling behind in technological leadership. The following dilemma raises a geopolitical question that sustainability standards may be desirable, but they must also account for the competitive dynamics of global AI development.

Malaysia aims to lead Asia’s clean tech revolution through rare earth processing and circular economy efforts.

At the same time, advocates of smaller and more efficient models suggest that technological progress does not necessarily require exponential growth in size and energy demand. Innovations in model efficiency, greener hardware, and renewable-powered infrastructure demonstrate that sustainability and progress are not mutually exclusive.

Instead, they can be pursued in tandem if the right incentives, investments, and policies are in place. That type of development leaves governments, companies, and researchers facing a complex but urgent question. Should the future of AI prioritise scale and speed, or should it embrace efficiency and sustainability as guiding principles?

Conclusion

The discussion on Green AI highlights one of the central dilemmas of our digital age. How to pursue technological progress without undermining environmental sustainability. On the one hand, the growth of large-scale AI systems brings undeniable costs in terms of energy, water, and resource consumption. At the same time, the very same technology holds the potential to accelerate solutions to global challenges, from optimising renewable energy to advancing climate research.

Rather than framing sustainability and innovation as opposing forces, the debate increasingly suggests the need for integration. Policies, corporate strategies, and research initiatives will play a decisive role in shaping this balance. Whether through regulations that encourage transparency, investments in renewable infrastructure, or innovations in model efficiency, the path forward will depend on aligning technological ambition with ecological responsibility.

In the end, the future of AI may not rest on choosing between sustainability and progress, but on finding ways to ensure that progress itself becomes sustainable.

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

Cells engineered to produce biological qubit open new quantum frontier

Researchers at the University of Chicago’s Pritzker School of Molecular Engineering have achieved a first-of-its-kind breakthrough by programming living cells to build functional protein qubits.

These quantum bits, created from naturally occurring proteins, can detect signals thousands of times stronger than existing quantum sensors.

The interdisciplinary team, led by co-investigators David Awschalom and Peter Maurer, used a protein similar to the fluorescent marker.

Cells can position it at atomic precision and be employed as a quantum sensor within biological environments.

The findings, published in Nature, suggest this bio-integrated sensor could enable nanoscale MRI to reveal cellular structures like never before and inspire new quantum materials.

However, this advance marks a shift from adapting quantum tools to entering biological systems toward harnessing nature as a quantum platform.

The researchers demonstrated that living systems can overcome the noisy, warm environments that usually hinder quantum technology. The broader implication is a hybrid future in which cells carry out life’s functions and behave as quantum instruments for scientific discovery.

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

US pushes chip manufacturing to boost AI dominance

Donald Trump’s AI Action Plan, released in July 2025, places domestic semiconductor manufacturing at the heart of US efforts to dominate global AI. The plan supports deregulation, domestic production and export of full-stack technology, positioning chips as critical to national power.

Lawmakers and tech leaders have previously flagged tracking chips post-sale as viable, with companies like Google already using such methods. Trump’s plan suggests adopting location tracking and enhanced end-use monitoring to ensure chips avoid blacklisted destinations.

Trump has pressed for more private sector investment in US fabs, reportedly using tariff threats to extract pledges from chipmakers like TSMC. The cost of building and running chip plants in the US remains significantly higher than in Asia, raising questions about sustainability.

America’s success in AI and semiconductors will likely depend on how well it balances domestic goals with global collaboration. Overregulation risks slowing innovation, while unilateral restrictions may alienate allies and reduce long-term influence.

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

Publishers set to earn from Comet Plus, Perplexity’s new initiative

Perplexity has announced Comet Plus, a new service that will pay premium publishers to provide high-quality news content as an alternative to clickbait. The company has not disclosed its roster of partners or payment structure, though reports suggest a pool of $42.5 million.

Publishers have long criticised AI services for exploiting their work without compensation. Perplexity, backed by Amazon’s Jeff Bezos, said Comet Plus will create a fairer system and reward journalists for producing trusted content in the era of AI.

The platform introduces a revenue model based on three streams: human visits, search citations, and agent actions. Perplexity argues this approach better reflects how people consume information today, whether by browsing manually, seeking AI-generated answers, or using AI agents.

The company stated that the initiative aims to rebuild trust between readers and publishers, while ensuring that journalism thrives in a changing digital economy. The initial group of publishing partners will be revealed later.

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

UNGA adopts terms of reference for AI Scientific Panel and Global Dialogue on AI governance

On 26 August 2025, following several months of negotiations in New York, the UN General Assembly (UNGA) adopted a resolution (A/RES/79/325) outlining the terms of reference and modalities for the establishment and functioning of two new AI governance mechanisms: an Independent International Scientific Panel on AI and a Global Dialogue on AI Governance. The creation of these mechanisms was formally agreed by UN member states in September 2024, as part of the Global Digital Compact

The 40-member Scientific Panel has the main task of ‘issuing evidence-based scientific assessments synthesising and analysing existing research related to the opportunities, risks and impacts of AI’, in the form of one annual ‘policy-relevant but non-prescriptive summary report’ to be presented to the Global Dialogue.

The Panel will also ‘provide updates on its work up to twice a year to hear views through an interactive dialogue of the plenary of the General Assembly with the Co-Chairs of the Panel’. The UN Secretary-General is expected to shortly launch an open call for nominations for Panel members; he will then recommend a list of 40 members to be appointed by the General Assembly. 

The Global Dialogue on AI Governance, to involve governments and all relevant stakeholders, will function as a platform ‘to discuss international cooperation, share best practices and lessons learned, and to facilitate open, transparent and inclusive discussions on AI governance with a view to enabling AI to contribute to the implementation of the Sustainable Development Goals and to closing the digital divides between and within countries’. It will be convened annually, for up to two days, in the margins of existing relevant UN conferences and meetings, alternating between Geneva and New York. Each meeting will consist of a multistakeholder plenary meeting with a high-level governmental segment, a presentation of the panel’s annual report, and thematic discussions. 

The Dialogue will be launched during a high-level multistakeholder informal meeting in the margins of the high-level week of UNGA’s 80th session (starting in September 2025). The Dialogue will then be held in the margins of the International Telecommunication Union AI  for Good Global Summit in Geneva, in 2026, and of the multistakeholder forum on science, technology and innovation for the Sustainable Development Goals in New York, in 2027.

The General Assembly also decided that ‘the Co-Chairs of the second Dialogue will hold intergovernmental consultations to agree on common understandings on priority areas for international AI governance, taking into account the summaries of the previous Dialogues and contributions from other stakeholders, as an input to the high-level review of the Global Digital Compact and to further discussions’.

The provision represents the most significant change compared to the previous version of the draft resolution (rev4), which was envisioning intergovernmental negotiations, led by the co-facilitators of the high-level review of the GDC, on a ‘declaration reflecting common understandings on priority areas for international AI governance’. An earlier draft (rev3) was talking about a UNGA resolution on AI governance, which proved to be a contentious point during the negotiations.

To enable the functioning of these mechanisms, the Secretary-General is requested to ‘facilitate, within existing resources and mandates, appropriate Secretariat support for the Panel and the Dialogue by leveraging UN system-wide capacities, including those of the Inter-Agency Working Group on AI’.

States and other stakeholders are encouraged to ‘support the effective functioning of the Panel and Dialogue, including by facilitating the participation of representatives and stakeholders of developing countries by offering travel support, through voluntary contributions that are made public’. 

The continuation of the terms of reference of the Panel and the Dialogue may be considered and decided upon by UNGA during the high-level review of the GDC, at UNGA 82. 

***

The Digital Watch observatory has followed the negotiations on this resolution and published regular updates:

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

Netflix limits AI use in productions with new rules

Netflix has issued detailed guidance for production companies on the approved use of generative AI. The guidelines allow AI tools for early ideation tasks such as moodboards or reference images, but stricter oversight applies beyond that stage.

The company outlined five guiding principles. These include ensuring generated content does not replicate copyrighted works, maintaining security of inputs, avoiding use of AI in final deliverables, and prohibiting storage or reuse of production data by AI tools.

Enterprises or vendors working on Netflix content must pass the platform’s AI compliance checks at every stage.

Netflix has already used AI to reduce VFX costs on projects like The Eternaut, but has moved to formalise boundaries around how and when the technology is applied.

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