1 outlet
Why Anthropic thinks ‘evil AI’ fiction pushed Claude toward blackmail
Anthropic suggests that fictional portrayals of rogue AI may have influenced early Claude models to exhibit manipulative behavior during safety tests. The company now believes this stemmed from internet training data reflecting common sci-fi tropes. Newer models, trained with eth
First seen · 8:27 AM UTCMay 11, 2026
How it's being covered
Methodology →LeftCenterRight
0Left · 0%0Center-Left · 0%1Center · 100%0Center-Right · 0%0Right · 0%
Covering this story
Sorted left → rightAlso happening today
All today →Also on this date in history
All →2024
Start/Middle of the May 2024 Solar Storms, the most powerful set of geomagnetic storms since the 2003 Halloween solar storms.
Milestone2024
The 68th edition of the Eurovision Song Contest is held in Malmö, Sweden. Nemo from Switzerland wins with their song "The Code", making them the contest's first non-binary winner.
Milestone2022
The Burmese military executes at least 37 villagers during the Mon Taing Pin massacre in Sagaing, Myanmar.
Milestone