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
-
Eurovision 2026 opens amid Gaza war tensions
-
You Can Tell What Kind Of Day You’ll Have By Where You Fell Asleep The Night Before
-
5 Zodiac Signs Have Very Good Horoscopes On Thursday, May 14, 2026
-
5 Perks You Didn’t Realize Came With Ikea’s Free Family Membership
-
iPhone 17 Users Ineligible for $250 Million Siri Settlement Agreement of Apple
