L'Analyse Longue — AI safety & alignment

Nia et Clara discutent des actualités du jour.

Sources :

• LessWrong: Some Thoughts on Bengio's Scientist AI
• LessWrong: Donating 80% While It Still Counts
• LessWrong: Cognitive Security as an AI Safety Cause Area
• LessWrong: Improving Petri scheming audits with environment blueprints