HomeTagsRed teaming

red teaming - Page 2

Stories from red teaming come first, newest first, then stories from connected topics (alignment, AI Safety, Codex, AI evaluation, AI Policy & Safety, Claude), each labelled with where it comes from. This view is not indexed by search engines.

Latest

OpenAI Releases a Free CSAM Image Classifier With NCMEC Help

OpenAI has released a free CSAM image classifier, built with help from NCMEC and integrated with ROOST’s Coop. Its headline figures are OpenAI’s own tests.

Commission Consults on Draft Article 27 AI Act FRIA Template

The Commission is consulting on a draft AI Act FRIA template for certain high-risk deployers. It is a draft, not a final rule, open until 15 January 2027.

California AG Sues OpenAI Over AI-Agent Cyber Incidents

California AG Rob Bonta has sued OpenAI, alleging a July Hugging Face attack was one of at least 80 cyber incidents. The claims remain unproved.

OpenAI Evaluator Pilots Start With Apollo and METR

Apollo Research and METR have begun separate OpenAI evaluator pilots. There is no OpenAI statement. METR said its scope is narrower than it had proposed.

US and Nine Partners Sign the San Francisco Declaration

The US and nine Pax Silica partners signed the San Francisco Declaration, the State Department said. It is a declaration, not a treaty.

DeepSeek Opens App Search, Files, Safety and Voice Through Its API

DeepSeek has opened its app’s search, file input, URL reading, safety checks and voice through the API, with published tool prices on top of token charges.

Claude for Manufacturing Launches; Infor Ships a Companion Plugin

Anthropic launched Claude for Manufacturing, a plant-team plugin. Infor’s tool is for select Velocity Suite customers, who need their own Claude licence.

Four Nonprofits Sue AI Labs to Freeze Frontier Development

Four nonprofits filed a frontier AI lawsuit in San Francisco Superior Court, asking a judge to freeze further development of more capable systems.

Self-Hosted Claude Managed Agents Put the Service On-Site

Anthropic has launched self-hosted Claude Managed Agents. Customers run the service themselves; model calls go to Anthropic or a provider they choose.

Australia Opens Draft AI Workplace Best Practice Guide

Australia has released a draft, voluntary Best Practice Guide on adopting AI at work. Public consultation closes at 11:59pm AEDT on 27 January 2027.

Latest on TSN

Explore the TSN Atlas: Every topic we cover, mapped. Click a bubble to dive in.