Why Anthropic’s New AI Model Sometimes Tries to ‘Snitch’

28 mai 2025 | Kylie Robison
The internet freaked out after Anthropic revealed that Claude attempts to report “immoral” activity to authorities under certain conditions. But it’s not something users are likely to encounter.
 Site référencé:  Wired
 

Wired 

DOGE Is Busier Than Ever—and Trump Says Elon Musk Is 'Really Not Leaving'
30/05/2025
A Hacker May Have Deepfaked Trump’s Chief of Staff in a Phishing Campaign
30/05/2025
Republican Operatives Want to Distance Themselves From Elon Musk's DOGE
30/05/2025
How the Loudest Voices in AI Went From ‘Regulate Us’ to ‘Unleash Us’
30/05/2025
Cops in Germany Claim They’ve ID’d the Mysterious Trickbot Ransomware Kingpin
30/05/2025
Fur Trim System Review : Simple, Pain-Free Body Hair Removal
30/05/2025