Ctrl+AI+Reg - 29 June 2026
Your shortcut to AI regulation, law and policy updates around the world.
AI Regulation Updates
In this issue:
Updates from Pax Silica (x2), Five Eyes, UK, US, Japan, China, South Korea, Australia.
Opinion on tokenomics and tokenpolitics, and 3 reflections in my day to day life.
Weekly Trend Analysis (for paid subscribers).
See fuller updates in the ‘Global Updates’ tab in the Global AI Regulation Tracker (English version | Chinese version)
Previous update
Global
🌐 [25 June 2026] Outcomes of the Second Pax Silica Summit: The US State Department has reported that the Pax Silica initiative, launched in December 2025, has achieved significant outcomes at its second summit, including the signing of a Joint Statement on AI Opportunity. Additionally, ten new signatories—Argentina, Chile, Costa Rica, El Salvador, the European Union, Germany, Greece, Kazakhstan, the Netherlands, and Panama—joined Pax Silica, bringing total signatories to 24. Related updates:
🇬🇧 🇺🇸 🇨🇦 🇦🇺 🇳🇿 [23 June 2026] Five Eyes agencies call for urgent action to manage AI risks: The Five Eyes cyber security agency leaders (from Australia, Canada, US, UK and New Zealand) have released a joint statement emphasising that the rapidly evolving landscape of AI is fundamentally transforming cyber risk, with frontier AI models expected to surpass current industry expectations within months, significantly enhancing both offensive and defensive cyber capabilities.
Europe
🇬🇧 [24 June 2026] FCA chief Rathi calls for regulatory rethink on AI evolution: Speaking at the Agents of Change: Generative and Agentic AI in Financial Services 2026 conference in London, Nikhil Rathi, Chief Executive of the UK’s Financial Conduct Authority (FCA), stated that the rapid evolution of AI is outpacing existing regulatory frameworks, necessitating a fundamental rethinking of regulatory approaches. Rathi emphasized that traditional rule-making is insufficient in certain areas due to the swift advancements in AI technology. The FCA is exploring the use of agentic AI to enhance its supervisory capabilities, particularly in monitoring wholesale markets, by leveraging large datasets and supervisory judgment to detect market abuse more efficiently.
Americas
🇺🇸 [25 June 2026] US government to decide AI technology access: It is reported that the White House’s Office of the National Cyber Director (ONCD) and Office of Science and Technology Policy (OSTP) has issued a formal request to OpenAI directing it to limit the initial commercial rollout of its GPT-5.6 model family — comprising Sol (a flagship agentic model with advanced capabilities in coding, biology and cybersecurity), Terra (a general-purpose model), and Luna (a fast, lower-cost model) — to approximately 20 government-vetted commercial partners, rather than conducting a general public release. The directive invoked the voluntary frontier model pre-release access framework established under Executive Order 14409, “Promoting Advanced Artificial Intelligence Innovation and Security,” signed by President Trump on 2 June 2026.
🇺🇸 [24 June 2026] US House subcommittee advances bill to protect consumers from AI energy costs: The US House subcommittee has advanced bipartisan legislation “Protecting Families from AI Data Center Energy Costs Act” aimed at shielding American consumers from electricity rate hikes attributable to AI infrastructure, with the legislation including specific provisions to prevent increased energy costs resulting from AI development and deployment.
🇺🇸 [24 June 2026] AI Data Center Moratorium Act introduced: The House of Representatives has introduced the AI Data Center Moratorium Act, with Representative Alexandria Ocasio-Cortez (NY-14) leading the legislation, and the Senate companion led by Senator Bernie Sanders (I-VT). The Act proposes a temporary federal moratorium on the construction of new data centers and the expansion of existing data centers until Congress enacts comprehensive legislation to regulate AI’s economic, environmental, and safety impacts.
Asia
🇯🇵 [26 June 2026] Japan lower house adopts social media election rules, covering AI generated content: It is reported that Japanese lawmakers in the lower house approved a bill aimed at curbing false and misleading election-related information, including AI-generated content, that could influence elections.
🇨🇳 [23 June 2026] State Administration for Market Regulation promotes high-quality development of China’s advertising industry: The State Administration for Market Regulation has published a joint opinion that outlines a comprehensive plan to optimize the industry’s spatial layout through regional collaboration and specialized digital clusters, accelerate the digital transformation of traditional advertising, and foster innovative technologies like AI, big data, and blockchain.
🇰🇷 [23 June 2026] KFTC announces proposed amendment to labeling and advertising rules: South Korea’s Fair Trade Commission (KFTC) announced an administrative notice of proposed amendment to the “Operation of Substantiation of Labeling and Advertising,” with a public comment period from June 23, 2026, to July 13, 2026, aiming to enhance the regulation of advertising substantiation, particularly for new technology products such as AI, by clarifying the obligation for prior verification when advertising AI functions and related claims such as “improving concentration and memory,” “ingredients harmless to the human body,” “feather XX%,” and “No. 1 in academic performance improvement,” which require significant verification regarding human health, safety, and performance based on previous rulings. The amendments also include the preparation and presentation of checklists for self-verification by businesses to improve predictability and operational clarity.
Oceania
🇦🇺 [26 June 2026] Another three ‘nudifying’ services withdraw from Australia as regulatory pressure mounts over child safety concerns: The eSafety Commissioner has enforced regulatory actions against three AI-powered “nudifying” services widely used by Australians, which have subsequently withdrawn their services from the country following enforcement under Australia’s Age-Restricted Material codes, designed to prevent children under 18 from accessing or generating age-restricted content, including sexually explicit material.
🇦🇺 [25 June 2026] Australian regulator warns banks of urgent frontier AI risks: The Australian Prudential Regulation Authority (APRA) has issued a warning that frontier AI models present an urgent and escalating risk to the financial sector, emphasizing the need for enhanced information sharing among regulated entities.
🇦🇺 [23 June 2026] Senator David Pocock alleges secret AI copyright deal under consideration: The ABC reports that federal politician David Pocock has publicly alleged that the Albanese government is confidentially considering two competing proposals to modify Australia’s copyright policy concerning AI, specifically regarding the training of AI systems on Australian copyright material, a claim that Industry and Science Minister Tim Ayres dismissed as “reckless speculation” in Parliament, although he did not deny or reject the existence of such cabinet considerations.
Opinion
See my post about tokenomics and tokenpolitics, and 3 reflections in my day to day life.
Weekly Trend Analysis
For the monthly trend analysis, check out the below article
The story of the past week is one of contradiction.
In public, the United States is making the case for openness. Its Pax Silica initiative has grown to twenty-four signatories, and a new joint statement speaks of opportunity, shared supply chains, and private investment in the infrastructure AI depends on, all framed as competition with China.
If that were the whole picture, you would expect American policy to push models out into the market as fast as possible. Instead, the same government did the opposite with its most direct action of the week.