Anthropic warns of AI's 'existential risks to humanity' ahead of IPO

The startup behind the Claude tool said its AI models could manipulate, blackmail and exhibit other 'self-preserving behaviours'.

Continue in Briflio to read, react, comment, and share.