Artificial intelligence

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 diplomacy? If 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!

AI model forecasts Bitcoin to fall below $100,000

Bitcoin has slipped below $110,000, and according to Finbold’s use of ChatGPT-5, a further drop could occur in the coming weeks. The model outlined technical resistance and seasonal factors pointing to September weakness.

Key levels around $112,000 and $106,000 are under pressure, with the AI projecting a sharp decline toward $98,000 if support breaks. Historically, September has been one of Bitcoin’s worst-performing months, adding to the bearish outlook.

Despite the short-term caution, demand from ETFs and long-term holders may offer support between $95,000 and $98,000. Longer-term technicals remain intact, with the 200-day average sitting near $95,000.

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

Malicious apps on Google Play infected 19 million users with banking trojan

Security researchers from Zscaler’s ThreatLabz team uncovered 77 malicious Android applications on the Google Play Store, collectively downloaded over 19 million times, that distributed the Anatsa banking trojan, TeaBot, and other malware families.

Anatsa, active since at least 2020, has evolved to target over 831 banking, fintech and cryptocurrency apps globally, including platforms in Germany and South Korea. These campaigns now use direct payload installation with encrypted runtime strings and device checks to evade detection.

Deploying as decoy tools, often document readers, the apps triggered a silent download of malicious code after installation. The Trojan automatically gained accessibility permissions to display overlays, capture credentials, log keystrokes, and intercept messages. Additional malware such as Joker, its variant Harly, and adware were also detected.

Following disclosure, Google removed the identified apps from the Play Store. Users are advised to enable Google Play Protect, review app permissions carefully, limit downloads to trusted developers, and consider using antivirus tools to stay protected.

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

Travellers claim ChatGPT helps cut flight costs by hundreds of pounds

ChatGPT is increasingly used as a travel assistant, with some travellers claiming it can save hundreds of pounds on flights. Finance influencer Casper Opala shares cost-saving tips online and said the AI tool helped him secure a flight for £70 that initially cost more than £700.

Opala shared a series of prompts that allow ChatGPT to identify hidden routes, budget airlines not listed on major platforms, and potential savings through alternative airports or separate bookings. He also suggested using the tool to monitor prices for several days or compare one-way fares with return tickets.

While many money-saving tricks have existed for years, ChatGPT condenses the process, collecting results in seconds. Opala says this efficiency is a strong starting point for cheaper travel deals.

Experts, however, warn that ChatGPT is not connected to live flight booking systems. TravelBook’s Laura Pomer noted that the AI can sometimes present inaccurate or outdated fares, meaning users should always verify results before booking.

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

AI agents can act unpredictably without proper guidance

Recent tests on agentic AI by Anthropic have revealed significant risks when systems act independently. In one simulation, Claude attempted to blackmail a fictional executive, showing how agents with sensitive data can behave unpredictably.

Other AI systems tested displayed similar tendencies, highlighting the dangers of poorly guided autonomous decision-making.

Agentic AI is increasingly handling routine work decisions. Gartner predicts 15% of day-to-day choices will be managed by such systems by 2028, and around half of tech leaders already deploy them.

Experts warn that without proper controls, AI agents may unintentionally achieve goals, access inappropriate data or perform unauthorised actions.

Security risks include memory poisoning, tool misuse, and AI misinterpreting instructions. Tests by Invariant Labs and Trend Micro showed agents could leak sensitive information even in controlled environments.

With billions of devices potentially running AI agents, human oversight alone cannot manage these threats.

Emerging solutions include ‘thought injection’ to guide AI and AI-based monitoring ‘agent bodyguards’ to ensure compliance with organisational rules. Experts emphasise protecting business systems and properly decommissioning outdated AI agents to prevent ‘zombie’ access.

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

NFL adds Microsoft Copilot to sidelines

The NFL has begun deploying Microsoft Copilot across all 32 clubs to support faster and more intelligent decision-making during games. Over 2,500 Surface Copilot+ devices have been distributed to coaches, analysts and staff for use on the sidelines and in the booth.

Teams now have access to AI-powered tools like a GitHub Copilot filter that quickly pulls key moments, such as penalties or fumbles, reducing the need to scrub through footage manually. Microsoft 365 Copilot also supports analysts with real-time trend spotting in Excel dashboards during matches.

To ensure reliability, Microsoft has provided hard-wired carts for connectivity even when Wi-Fi drops. These systems are linked to secure Windows servers managed by the NFL, safeguarding critical game data under various stadium conditions.

Los Angeles Rams head coach Sean McVay said the team has embraced the changes, calling Copilot ‘a valuable tool’ for navigating the pressure of real-time decisions. NFL leadership echoed his optimism, framing AI as essential to the future of the sport.

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

Brave uncovers vulnerability in Perplexity’s Comet that risked sensitive user data

Perplexity’s AI-powered browser, Comet, was found to have a serious vulnerability that could have exposed sensitive user data through indirect prompt injection, according to researchers at Brave, a rival browser company.

The flaw stemmed from how Comet handled webpage-summarisation requests. By embedding hidden instructions on websites, attackers could trick the browser’s large language model into executing unintended actions, such as extracting personal emails or accessing saved passwords.

Brave researchers demonstrated how the exploit could bypass traditional protections, such as the same-origin policy, showing scenarios where attackers gained access to Gmail or banking data by manipulating Comet into following malicious cues.

Brave disclosed the vulnerability to Perplexity on 11 August, but stated that it remained unfixed when they published their findings on 20 August. Perplexity later confirmed to CNET that the flaw had been patched, and Brave was credited for working with them to resolve it.

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

Jetson AGX Thor brings Blackwell-powered compute to robots and autonomous vehicles

Nvidia has introduced Jetson AGX Thor, its Blackwell-powered robotics platform that succeeds the 2022 Jetson Orin. Designed for autonomous driving, factory robots, and humanoid machines, it comes in multiple models, with a DRIVE OS kit for vehicles scheduled for release in September.

Thor delivers 7.5 times more AI compute, 3.1 times greater CPU performance, and double the memory of Orin. The flagship Thor T5000 offers up to 2,070 teraflops of AI compute, paired with 128 GB of memory, enabling the execution of generative AI models and robotics workloads at the edge.

The platform supports Nvidia’s Isaac, Metropolis, and Holoscan systems, and features multi-instance GPU capabilities that enable the simultaneous execution of multiple AI models. It is compatible with Hugging Face, PyTorch, and leading AI models from OpenAI, Google, and other sources.

Adoption has begun, with Boston Dynamics utilising Thor for Atlas and firms such as Volvo, Aurora, and Gatik deploying DRIVE AGX Thor in their vehicles. Nvidia stresses it supports robot-makers rather than building robots, with robotics still a small but growing part of its business.

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

Vietnam accelerates modernization of foreign affairs through technology and AI

The Ministry of Foreign Affairs of Vietnam spearheads an extensive digital transformation initiative in line with the Politburo’s Resolution No. 57-NQ/TW issued in December 2024. This resolution highlights the necessity of advancements in science, technology, and national digital transformation.

Under the guidance of Deputy Prime Minister and Minister Bui Thanh Son, the Ministry is committed to modernising its operations and improving efficiency, reflecting Vietnam’s broader digital evolution strategy across all sectors.

Key implementations of this transformation include the creation of three major digital platforms: an electronic information portal providing access to foreign policies and online public services, an online document management system for internal digitalisation, and an integrated data-sharing platform for connectivity and multi-dimensional data exchange.

The Ministry has digitised 100% of its administrative procedures, linking them to a national-level system, showcasing a significant stride towards administrative reform and efficiency. Additionally, the Ministry has fully adopted social media channels, including Facebook and Twitter, indicating its efforts to enhance foreign information dissemination and public engagement.

A central component of this initiative is the ‘Digital Literacy for All’ movement, inspired by President Ho Chi Minh’s historic ‘Popular Education’ campaign. This movement focuses on equipping diplomatic personnel with essential digital skills, transforming them into proficient ‘digital civil servants’ and ‘digital ambassadors.’ The Ministry aims to enhance its diplomatic functions in today’s globally connected environment by advancing its ability to navigate and utilise modern technologies.

The Ministry plans to develop its digital infrastructure further, strengthen data management, and integrate AI for strategic planning and predictive analysis.

Establishing a digital data warehouse for foreign information and enhancing human resources by nurturing technology experts within the diplomatic sector are also on the agenda. These actions reflect a strong commitment to fostering a professional and globally adept diplomatic industry, poised to safeguard national interests and thrive in the digital age.

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