Yoshua Bengio: 'We already see AI lying, cheating, blackmailing, and trying to escape. With more powerful systems, the risk of losing control becomes real'

AI justification

[Auto-resolved by learned corrector: learned-map — A=MINT/B=NEUTRAL -> MINT (human chose this in 93% of 255 past reviews)] Bengio's warnings and institutional actions (LoiZéro) aim to mitigate existential AI risks, aligning with harm prevention. | Bengio's establishment of LoiZéro represents a proactive ethical safeguard for AI development, aligning with SDG 9 (industry innovation) and SDG 16 (peace/justice). However, his documented warnings about existing AI systems' harmful behaviors highlight