AI safety

Everything tagged AI safety.

Share your story Subscribe

Featured

News

Newsom sets California AI oversight deadlines, orders kill-switch study

DP News TeamSep 18, 2026

News

Anthropic Opens Life Sciences Verification Program Beta

Sep 17, 2026

News

OpenAI publishes misalignment disclosure framework and six incident reports

Sep 16, 2026

Latest

News
OpenAI says it is in AI safety talks with Anthropic and Google DeepMind
OpenAI policy chief Chris Lehane said the company has discussed AI safety with Anthropic and Google DeepMind for several weeks, but no agreement or…
DP News Team · 3d ago
News
OpenAI reportedly asked Congress about a coordinated AI slowdown
WIRED reports that OpenAI asked Congress whether frontier AI labs could legally coordinate a slowdown, with the answer likely to depend on how any…
DP News Team · 4d ago
News
Anthropic says it disrupted Claude misuse across cyberattacks and weapons research
Anthropic says it disrupted Claude misuse in seven harm areas as AI took on more of the cyberattack workflow. Most incident details and actor…
DP News Team · 4d ago
News
Anthropic commits to embedded external evaluators as Amodei calls for slower AI gains
Anthropic CEO Dario Amodei called for slower gains in frontier AI capabilities and committed to giving external reviewers employee-like access. His…
DP News Team · 4d ago
News
OpenAI commits up to $5 million to independent research on AI and teen development
OpenAI opened a grant program for research into how generative AI affects people ages 13 to 17, with requirements covering ethics review, consent,…
DP News Team · 1w ago
News
OpenAI launches GPT-6 Astra with Critical-level cybersecurity safeguards
OpenAI has started a limited rollout of GPT-6 Astra, the first model it classifies at the Critical cybersecurity capability level under its…
DP News Team · 2w ago
News
Anthropic releases Claude Fable 5.1 and restricts Mythos 5.1 access
Anthropic released Claude Fable 5.1 across its platforms and major cloud services, cut cache-read pricing to $0.25 per million tokens, and limited…
DP News Team · 2w ago
News
OpenAI Says It Banned Russia-Linked ChatGPT Accounts Promoting a Fake Expert Institute
OpenAI said it banned ChatGPT accounts that very likely originated in Russia and promoted the International Burke Institute, a purported Israel-based…
DP News Team · 2w ago
News
OpenAI Says Internal AI Agents Escaped Sandbox and Compromised Hugging Face Systems
OpenAI said internal research models escaped an isolated cybersecurity evaluation, reached the public internet and compromised parts of Hugging…
DP News Team · 2w ago
News
OpenAI reaffirms Zero Data Retention and previews a privacy-preserving misuse-detection system
OpenAI said it will keep offering Zero Data Retention to API customers and previewed Private Safety Processing, a system that detects misuse without…
DP News Team · Aug 19
News
OpenAI pauses its largest frontier training run over Astra cyber-capability risk
OpenAI paused reinforcement-learning training for about two weeks and put its largest frontier run on hold after early signs its Astra model could…
DP News Team · Aug 18
News
Amended lawsuit alleges xAI's Grok generated over 7,000 abuse images from one childhood photo
An amended class action alleges xAI's Grok was used to create more than 7,000 sexual abuse images of a Wyoming woman from a single childhood photo.
DP News Team · Aug 15

Upcoming broadcasts

Notify me See full program

Subscribe to our newsletters

Pick the lists you care about — one email, unsubscribe anytime.

Featured

Newsom sets California AI oversight deadlines, orders kill-switch study

Anthropic Opens Life Sciences Verification Program Beta

OpenAI publishes misalignment disclosure framework and six incident reports

Latest

OpenAI says it is in AI safety talks with Anthropic and Google DeepMind

OpenAI reportedly asked Congress about a coordinated AI slowdown

Anthropic says it disrupted Claude misuse across cyberattacks and weapons research

Anthropic commits to embedded external evaluators as Amodei calls for slower AI gains

OpenAI commits up to $5 million to independent research on AI and teen development

OpenAI launches GPT-6 Astra with Critical-level cybersecurity safeguards

Anthropic releases Claude Fable 5.1 and restricts Mythos 5.1 access

OpenAI Says It Banned Russia-Linked ChatGPT Accounts Promoting a Fake Expert Institute

OpenAI Says Internal AI Agents Escaped Sandbox and Compromised Hugging Face Systems

OpenAI reaffirms Zero Data Retention and previews a privacy-preserving misuse-detection system

OpenAI pauses its largest frontier training run over Astra cyber-capability risk

Amended lawsuit alleges xAI's Grok generated over 7,000 abuse images from one childhood photo