AI Research
Anthropic Let Researchers See Claude Usage, But Not Conversations
Each new query would have required a fresh privacy review, so partners tested their questions against a different, public dataset instead.
5d ago
#transparency
AI Safety
Waymo's NHTSA Child-Collision Answers Are Public, But Blacked Out
"Hands over documents" reads as disclosure. Handing material to a regulator and disclosing it to the public are different events, and only the first one happened.
Aug 22, 2026
AI Research
Anthropic's Own Method, Applied to Donated Chat Logs, Discards Half the Conversations
An independent project re-ran a lab's work-focused classification and found the filtered-out half is where the health, harassment and sexual content lives.
Aug 19, 2026
EU AI Act
About 190 Signed EU's AI Transparency Code, But Lists Total 235
Anthropic, Google, Meta, Microsoft, OpenAI and Mistral are on the provider list. The counting is confusing because companies sign both sections — and the Commission's own pages disagree.
Aug 1, 2026
EU AI Act
EU Starts Enforcing AI Act Sunday, But Most Is Deferred
Article 50 transparency duties apply from 2 August and the AI Office gains powers over general-purpose model providers. The high-risk regime that the argument was about is not among them.
Aug 1, 2026
AI Safety
Hugging Face Flew to San Francisco and Asked OpenAI for the Logs
Clément Delangue wants the activity traces from the rogue agents released for public study, plus $100 million in compute. OpenAI confirmed the meeting and promised a technical report.
Jul 27, 2026
EU AI Act
EU Publishes AI Act Transparency Guidelines, Two Weeks Before the Rules Take Effect
The European Commission issued final guidance on Article 50, spelling out how providers must label AI systems and synthetic content and how deployers must disclose deepfakes — ahead of the August 2 application date.
Jul 20, 2026
