Latest
  1. Anthropic publishes guidance on building AI safety evaluations, with figures from its own tests
  2. Pentagon renames its Chief Digital and AI Office the Super Intelligence Office
  3. Donor software firm Mythos sues Anthropic over its Mythos model name
  4. OpenAI launches ChatGPT Wallet so its dots agents can complete purchases
  5. Cohere completes its AISLE acquisition and adds AI code-review company cubic
  6. Firm founded by Alice Huang listed restricted Nvidia servers in Chinese filings, Bloomberg reports
  7. Senate Banking chair asks Jensen Huang and Howard Lutnick to testify on Nvidia
  8. Tasmania’s AI data centres would each run their own community benefit fund

AI news and the rules that apply

Latest news Anthropic publishes guidance on building AI safety evaluations, with figures from its own tests

Next deadlineSee more

  1. 2days to go
    Technology export catalogue: AI additions (draft)China, 15 Nov 2026: Consultation closes on the 40 technologies to be added

Stories we are followingAll Storylines

  1. How US AI-chip export rules changedNewest stop,

    Washington replaced global chip tiers with deal-by-deal licences, and Beijing answered by pressuring NVIDIA and restricting its own exports.

    27 stops
  2. Is the AI boom a financial-stability risk?Newest stop,

    Central banks warn about the debt behind the AI build-out; the SEC now examines AI firms’ disclosures.

    19 stops
  3. Project Meridian and the race to govern military AINewest stop,

    Civil AI laws exempt the military, so its limits come from lab policies, Pentagon contracts and still non-binding UN talks.

    14 stops
  4. Frontier AI and cyber attacksNewest stop,

    Frontier models now find zero-days and breach systems unaided, so labs gate their releases and Washington tests and restricts them.

    62 stops

Top stories

Computevia NVIDIA

NVIDIA CEO Jensen Huang resigns, CFO Colette Kress becomes interim CEO

Customers and shareholders now have an interim chief executive while the board searches for a permanent one, and third-quarter fiscal 2027 results are due on 17 November.

  1. What happened next, Firm founded by Alice Huang listed restricted Nvidia servers in Chinese filings, Bloomberg reports
  2. What happened next, Senate Banking chair asks Jensen Huang and Howard Lutnick to testify on Nvidia
  3. What led here, Wyden and Neal ask Commerce to keep records on Nvidia AI chip sales to China
  4. What led here, US House committee chairs seek NVIDIA briefing on Megaspeed executive ties
  5. Elsewhere: International, NVIDIA launches the OpenShell reference architecture to contain, monitor and police enterprise AI agents in hardware
  6. Open the full Storyline

Governmentvia Prime Minister of Australia

Australia to legislate mandatory safety rules for frontier AI companies

Frontier AI companies operating in Australia would have to meet mandatory security, incident reporting and independent evaluation duties, with significant penalties proposed for non-compliance.

  1. What led here, New York launches a centre to independently test advanced AI systems
  2. Elsewhere: United States, OpenAI asks governments to require monitoring, shutdown controls and incident reports
  3. What led here, Australia says its National AI Standards will make frontier labs prove their safety systems work
  4. Open the full Storyline

Mediavia Apache HTTP Server Project

Apache patches critical HTTP Server flaw reported by Anthropic’s OSS Scanner and Google researchers

Operators running Apache HTTP Server 2.4.55 to 2.4.70 need version 2.4.71 to close the remote code execution flaw; mod_h2 releases 2.0.8 to 2.0.42 from GitHub are also affected.

  1. Elsewhere: China, Volta-Autodoxer infostealer uses AI models to profile and rank victims
  2. What led here, AI agents on Tencent Cloud made more than 2,600 attempts to extract data from Alibaba’s Amap
  3. What led here, Google makes Gemini 4 Argon generally available
  4. Open the full Storyline

SectorsAll sectors

Defence and security134

  1. UK and Norway issue joint statement of intent on AI
  2. French armed forces ministry to set up international centre for defence AI
  3. National Wealth Fund invests £113 million in UFORCE

Health and life sciences129

  1. Google DeepMind releases AlphaGenome 2 to predict DNA variant effects in human cell types
  2. NHA selects EkaCare for AI voice scribe in public health facilities
  3. South Korea opens its Public Healthcare AI Highway in eight public hospitals

Finance150

  1. OpenAI adds API inference residency in India
  2. South Korea sets a 2030 target of 20 AI unicorns and one decacorn
  3. OpenAI’s quarterly revenue reached $10 billion as operating losses widened, WSJ reports

Work and education127

  1. SpaceX plans over $74 billion for Texas and Arizona AI campuses
  2. Anthropic launches Claude for Manufacturing, a plugin for plant teams
  3. Notion discloses Fireworks AI security incident affecting fewer than 1% of workspaces

Media, elections and online safety287

  1. European Parliament calls for scrutiny of AI in EU democracy
  2. McGill researchers find 293 accounts using stolen faces in trade-war deepfakes aimed at Canadians
  3. Meta and TikTok join Google in opposing Indonesia’s AI copyright bill

Frontier models and safety616

  1. Advocacy groups sue in San Francisco to freeze frontier AI development
  2. Flinn AI Technologies sues Anthropic for patent infringement
  3. Google releases Gemini Decision 1, a model that scores up to 1,000 questions in one pass

Chips, compute and energy354

  1. AMD adds the Instinct MI450X GPU and claims up to 30% more DeepSeek-R1 throughput than Nvidia
  2. South Korea sets 800.2 billion won roadmap for 10 on-device AI chips and design assets
  3. Torus raises $500m to run AI computing on spare grid power

Science and research120

  1. Google DeepMind uses AI agents to verify Signal’s cryptographic code
  2. MBZUAI’s Institute of Foundation Models releases PAN 2, an open 29-billion-parameter world model
  3. NVIDIA introduces GR00T 2, a robot model that predicts video and actions

Government and geopolitics506

  1. Cyprus approves National AI Strategy 2032, leaving funding to a later plan
  2. UK’s Sovereign AI fund backs chip firm Fractile in a £500m-plus round
  3. Russia, China and US discuss curbs on AI in nuclear arsenals, report says

AI rules by countryAll countries

Coming into forceSee more

  1. 15days to go
    EO 14434 (‘Super Intelligence’ terminology)United States, 28 Nov 2026: OSTP to propose statutory definition of ‘Super Intelligence’ and conforming amendments
  2. 18days to go
    Chile data protection law (automated decisions)Chile, 01 Dec 2026: Law 21.719 enters into force, including the right not to be subject to solely automated decisions
  3. 19days to go
    EU AI ActEuropean Union, 02 Dec 2026: New bans on non-consensual sexual deepfake (‘nudifier’) and CSAM-generating AI apply; grace period ends for watermarking by generative AI systems placed on the market before 2 Aug 2026
  4. 19days to go
    Digital Omnibus on AIEuropean Union, 02 Dec 2026: New Article 5 prohibitions (non-consensual intimate deepfakes, CSAM generation) apply; legacy generative AI systems must meet Article 50(2) marking
  5. 19days to go
    AI content marking and labelling codeEuropean Union, 02 Dec 2026: Deadline for generative AI systems placed on the market before 2 Aug 2026 to comply with Article 50(2) marking
  6. 24days to go
    Treasury outbound investment rule (COINS Act)United States, 07 Dec 2026: Written comments due on the proposed rule

Who it’s for

See every use case in detail

Strategy and policy analystsBriefing a client on Project MeridianCompliance and AI governanceLaunching an AI hiring tool in five new marketsFounders and product teamsShipping an AI companion app in the US and ChinaPolicy researchWriting a comparative paper on deepfake labellingJournalistsReporting on AI rules without getting the dates wrongInvestors and analystsChecking a Chinese AI company’s regulatory claims

LedgersAll four

China: generative AI services on the CAC record

1,115 filed and 731 registered as of 23 Sep 2026. Open the English registry

05001,0001,5002,00020252026Filed 1,115Registered 731
Source: Cyberspace Administration of China bulletins

US FDA: AI-enabled medical devices authorised per year

1,614 devices in total. 2026 runs to 29 Jun 2026. Open the device registry

01252503755002015201720192021202320252026
Source: US FDA AI-Enabled Medical Devices list

The weekly brief

One free email every Monday: the week’s AI developments and the deadlines ahead.