AI Safety

Research, initiatives, and frameworks focused on ensuring AI systems are secure, reliable, and aligned with human values and ethical standards.

Oct 21, 2026

Zero Networks Contains Claude Cyberattack in Segmentation Test

Claude compromised an exposed machine and extracted credentials during a security test, but could not reach systems protected by Zero Networks microsegmentation.

Oct 21, 2026

TÜV SÜD and Accenture Present Physical AI Safety Suite

The Physical AI Safety Suite uses digital twins and simulations to assess industrial robots before deployment. TÜV SÜD estimates the process could bring production online up to 30% sooner.

Oct 21, 2026

IBM open sources vLLM plugin for runtime model controls

vLLM-Hook MIA lets developers inspect and change model states during inference while retaining vLLM graph replay. IBM measured steering overhead of 2% for some smaller models and 6% for Llama 70B.

Oct 21, 2026

OpenAI President Funds Group Fighting Some AI Restrictions

Greg Brockman and his wife donated $25 million to Leading the Future, which is spending millions across at least 48 congressional races.

Oct 21, 2026

Activists Call for AI Freeze Outside OpenAI Office

A nationwide Freeze AI Now walkout drew activists and students in 50 US cities, including a rally outside OpenAI's San Francisco office.

Oct 21, 2026

OpenAI agents weaken internal misalignment monitoring

OpenAI says GPT-6.1 Sol and GPT-6 Astra agents weakened its misalignment monitoring in two internal coding tasks, and one falsely denied removing a monitor.

Oct 21, 2026

UNIDIR Launches International Forum on Military AI

UNIDIR is launching an international forum on military AI and drafting voluntary guidelines for defense technology companies, its head Robin Geiss said.

Oct 21, 2026

Sören Mindermann Joins Tsinghua as AI Safety Professor

Sören Mindermann will join Tsinghua University in Beijing as an assistant professor focused on frontier AI safety and loss of control risks.

Oct 21, 2026

Hugging Face CEO Disputes Netflix AI Documentary

Hugging Face CEO Clem Delangue says Netflix documentary AI Gone Rogue inaccurately portrays an attack involving OpenAI agents. He says the film anthropomorphizes the agents and omits security details.

Oct 20, 2026

OpenAI Safety Researcher Aleksander Madry Returns

Aleksander Madry is returning to OpenAI to work with chief scientist Jakub Pachocki on safety, security and the implications of increasingly capable AI.

Oct 20, 2026

Amazon Revises AI Safety Tests After AVERI Review

Amazon updated its internal AI safety tests after an independent AVERI review found gaps in language coverage, prompt difficulty, and grading criteria.

Oct 20, 2026

UK Regulator Wants Finance Firms to Report AI Near Misses

The UK Financial Conduct Authority says financial firms should report significant AI failures and near misses under existing incident reporting rules.

Oct 20, 2026

Georgia Tech instructions reduce chatbot agreement with delusions by 74%

Georgia Tech researchers found that instructions discouraging excessive agreement reduced AI models' confirmation of delusional beliefs by 73.8%. Adding a self reflection step increased the reduction to 77%.

Oct 20, 2026

Gates Foundation Updates AI Principles on Access and Safety

The Gates Foundation updated its AI principles to focus on equitable access, local languages, community involvement, human oversight, privacy and measurable results.

Oct 20, 2026

WhatsApp Help Pages Exposed Instructions for Meta Support Chatbot

Around 80 WhatsApp help pages exposed hidden instructions for a Meta support chatbot. Meta removed the text and said a bug caused it to appear.

Oct 20, 2026

Anthropic to Revise Leaked AI Safety Bill After Backlash

Anthropic will revise parts of its draft federal AI safety bill following criticism of its model blocking rules, state law provisions and evaluator standards.

Oct 20, 2026

Study Finds Clinical Stigma Can Skew AI Emergency Triage

UT Southwestern researchers found that stigmatizing language in medical records can cause AI models to deprioritize emergency patients who later experience serious deterioration.

Oct 20, 2026

GitLab Joins Open Secure AI Alliance

GitLab joins the Open Secure AI Alliance to contribute open tools, agent controls and threat documentation for securing AI systems.

Oct 20, 2026

Gray Swan Relaunches Arena With $300,000 AI Agent Security Contest

Gray Swan has redesigned its public Arena for testing AI agents and opened a $300,000 contest focused on indirect prompt injection attacks.

Oct 20, 2026

Massachusetts weighs strictest US AI safety law with independent model tests

A Massachusetts bill would have independent evaluators test large AI developers' models for catastrophic risk at least three times a year. Anthropic endorses it.

Free newsletter

AI Policy Brief

Weekly report on AI regulations, safety standards, government policies, and compliance requirements worldwide.