Artificial intelligence (AI) firm Anthropic says testing of its new system revealed it is sometimes willing to pursue "extremely harmful actions" such as attempting to blackmail engineers who say they will remove it.
Anthropic's new AI model turns to blackmail when engineers try to take it offline
-- TechCrunch (@techcrunch.com) May 22, 2025 at 1:56 PM
[image or embed]
Drudge Retort Headlines
US Senator Dragged Out of News Conference (133 comments)
Minnesota Democratic Lawmaker Killed after Multiple 'targeted' Shootings (113 comments)
Israel Hits Iran (94 comments)
Noem: Fed is 'liberating' Los Angeles (89 comments)
Huge Crowds at 'No Kings' Protests Across U.S. (82 comments)
Weather Could Impact Trump's Multi-Million Dollar Military Parade Saturday (46 comments)
DNC Ousts David Hogg as Vice-Chairman (26 comments)
US to Pull Some Personnel from the Middle East (24 comments)
Real Estate Market Crashing Before Our Eyes (22 comments)
No One Cares About Trump's Parade (20 comments)