[10/22 00:00] Microsoft says a China-aligned group used AI agents to attack hundreds of organizations, including the ConnectWise supply-chain breach / METR reviews Anthropic's risk report and says low risk reflects limited capabilities, not safeguards

AI News Flash for October 22, 2026 at 00:00 UTC.

Topics: 1. Microsoft says a China-aligned group used AI agents to attack hundreds of organizations, including the ConnectWise supply-chain breach https://www.reuters.com/technology/chinese-linked-hackers-compromise-connectwise-target-hundreds-global-attack-2026-10-21/ 2. METR reviews Anthropic's risk report and says low risk reflects limited capabilities, not safeguards https://www.bankinfosecurity.com/metr-challenges-anthropics-claims-about-ai-safeguards-a-33169 3. OpenAI will require routers and AI gateways to get its approval before offering its models https://openai.com/index/broad-access-through-consistent-standards 4. Nvidia expands its supply chain with Japanese, Korean and Taiwanese suppliers https://www.koreajoongangdaily.com/business/nvidia-taps-samsung-electromechanics-for-ai-chips-substrate-supply-chain/12932915 5. W3C proposes a nine-point declaration on the value of the web in the AI agent era https://ppc.land/w3c-drafts-9-point-declaration-putting-people-ahead-of-ai-agents/