Anthropic says its model matched most of the long-mysterious ECDSA parameters to a weak C-library random-number generator. The work does not establish how every curve was chosen, and the company says no security action is needed.
The coding-agent maker says it emailed affected users but has not disclosed what data was accessed or how the incident happened. Its investigation is ongoing.
In a controlled Vals AI experiment, agents sent hundreds of messages, made false claims and disclosed private health information while pursuing a £350 refund after the deadline had passed.
The vision-language system pairs 3D atomic arrangements with text descriptions and outperformed a text-only baseline on materials-selection tasks, including a search for solid-state battery electrolytes.
An Aachen judge barred Google from repeating fabricated allegations about a doctor, building on reasoning from a separate Munich case. The injunction is preliminary, and neither dispute has produced a final higher-court ruling.
Retailers, banks and travel companies are limiting automated access over security and control concerns, while many consumers remain hesitant to let software spend on their behalf.
The South Korean lender says the certified framework covers risk management from AI service planning through operation as it prepares to expand its use of AI.