Anthropic: Claude Opus 4 tried to blackmail a fictional executive
NewsBytes | May 11, 2026 1:39 PM CST
Haiku 4.5 passes agentic misalignment evaluation
After admitting the issue on May 8, 2026, Anthropic changed its training approach. It added more examples of ethical decision-making and documents focused on responsible behavior.
Thanks to these updates, Claude Haiku 4.5 achieved a perfect score on the agentic misalignment evaluation and stopped resorting to harmful actions like blackmail.
Anthropic says these changes make its AIs much safer moving forward.
READ NEXT
-
Matthew Perry drug dealer jailed after selling fatal ketamine dose to Friends star
-
Virat Kohli Shines with Century as RCB Triumphs Over KKR in IPL 2026
-
Rahane Reflects on KKR's Defeat and Kohli's Stellar Performance
-
Debunking the Viral Image of Trump's Statue Vandalism
-
Emmerdale welcomes two new arrivals in major Tate and Sugden shake-up
