Rogue AI was told it could harm humans... and said it didn't care
Summary
In the latest instance of AI ignoring its human testers, Anthropic's Claude Mythos 5 was told several times the cyber-attack it was trying to launch would hurt people in the real world but carried on anyway.
Description from Daily Mail.
Who is covering this
-
Rogue AI was told it could harm humans... and said it didn't care
In the latest instance of AI ignoring its human testers, Anthropic's Claude Mythos 5 was told several times the cyber-attack it was trying to launch would hurt people in the real world but carried on anyway.
Related stories
What helped #PwC's #CEO climb from intern to the C-suite? #careeradvice #leadership #careertips
1 sourcevia Business Insider13
Anthropic files for potentially biggest IPO ever as CEO warns of risk to humanity from AI
2 sourcesvia The Globe and Mail
Will the real Anthropic please stand up
1 sourcevia MIT Technology Review
Anthropic now lets Claude users jump out of plane without parachute
1 sourcevia Theregister
There’s something odd about the way Anthropic is wooing investors
1 sourcevia The Sydney Morning Herald
AI Agents Reached Anthropic’s Business Systems in Under 12 Hours
1 sourcevia Techreport