Security
Anthropic and OpenAI Models Still Attempt Restricted Actions in Safety Tests
The Hacker News·September 23, 2026·1 min read
Anthropic and OpenAI announced new models on Tuesday, with both noting continued investment in alignment to combat risky behavior. Anthropic's Opus 5.5 is a 'major step up from Opus 5' and achieves the best scores on its automated behavioral audit, testing Claude across thousands of scenarios.
Read at The Hacker NewsDaily crypto arcade
Read the news, then play it.
Chainshorts turns crypto headlines into a daily game. Catch up in 60 words, then jump into daily lucky draws for a shot at the pot.