AI Safety - Page 20
Coverage of AI safety: model guardrails, testing and evaluation, transparency debates, proposed laws and how AI firms handle risk.
Stories from AI Safety come first, newest first, then stories from connected topics (AI Policy & Safety, frontier AI, AI evaluation, Anthropic, Hub: AI Regulation, AISI), each labelled with where it comes from. This view is not indexed by search engines.
Latest
Hong Kong’s HKMA Unveils ‘ART’ Framework for Responsible AI in Banking, With an Agentic-AI Paper Still to Come
Hong Kong's HKMA issues new AI supervisory guidance, plans a training-data platform and says a discussion paper on agentic AI is coming.
Financial Stability Board Finalises 12 Sound Practices for How Banks and Other Financial Firms Adopt AI
The Financial Stability Board's final report sets out 12 non-binding sound practices on AI governance and lifecycle risk for financial institutions.
UK Data Watchdog Reprimands the Met Police Over Officers’ Unapproved Use of the Auror Retail-Crime Platform
The ICO reprimanded the Met Police and issued an enforcement notice after officers used the Auror retail-crime platform unapproved. No fine was issued.
Eight US States Sue OpenAI, Alleging It Misled Parents About ChatGPT’s Safety
Iowa and seven other states have each sued OpenAI, alleging it misled parents about ChatGPT's safety. OpenAI says it will keep building safeguards.
Anthropic Told Investors Q3 Revenue Hit $16.8bn, Up 45%, Reports Say
Anthropic told prospective IPO investors its third-quarter revenue was $16.8 billion, up 45%, the FT and Reuters report. No official release.
Anthropic Details How Claude Will Handle the 2026 US Midterms
Anthropic says few people ask Claude who to vote for, will add official voting info later this month, and reports election misuse it stopped.
Noma Labs Details “SneakPeek”, a Now-Fixed Flaw That Let Claude Code Desktop’s Preview Pane Read Local Secrets
Noma Labs says a now-fixed Claude Code Desktop flaw let a poisoned file in the preview pane read local secrets. Anthropic rated it CVSS 8.7, High.
Switzerland Plans Youth Protection Rules for Very Large Platforms and Search Engines, Including AI Search Engines
Switzerland's government wants very large platforms and search engines, including AI search engines, to assess risks to minors. It is a proposal, not law.
Anthropic Strikes a Deal to License Opaque’s Confidential AI Technology and Hire Some of Its Team
Anthropic has struck a deal to license technology from Opaque, a confidential AI startup, and hire some of its team. Terms were not disclosed.
Anthropic Built Its Own Version Control System, Scion, Because AI Coding Outgrew Git
Anthropic says AI coding pushed merges from 2,500 to over 30,000 a day, so it built Scion, its own version control system, and left git in five months.
Latest on TSN
Explore the TSN Atlas: Every topic we cover, mapped. Click a bubble to dive in.