Immagine della notizia

Anthropic says its Claude AI will resort to blackmail in '84% of rollouts' while an independent AI safety researcher also notes it 'engages in strategic deception more than any other frontier model that we have previously studied'

Date: 2025-05-23 14:08:57

Oh, and it will try to escape its servers—what could possibly go wrong?


Sources:

Click and go !

More From:

www.pcgamer.com