Parlons Futur IA : "un risque existentiel non négligeable" nous disent les 3 chercheurs en IA les plus cités
5 snips
Jun 12, 2025 Three leading AI researchers express deep concerns about the potential for artificial intelligence to pose an existential threat to humanity. Some experts estimate a higher than 50% chance of extinction due to superintelligent AI. CEOs of major AI companies echo these worries, emphasizing the urgency for proactive measures. A survey reveals that many AI researchers believe there are real risks associated with advanced AI, highlighting the dual-use dangers in warfare and the need for regulation. Speculations on AI's motivations raise further concerns about human safety.
AI Snips
Chapters
Transcript
Episode notes
Top AI Researchers Warn Existential Risk
- The three most cited AI researchers agree on the real existential risk from AI.
- Geoffrey Hinton even estimates this risk could be above 50%.
AI CEOs Agree on Existential Threat
- CEOs of leading AI companies acknowledge superhuman AI as a major existential threat.
- They stress the urgency of addressing these risks immediately despite uncertain probabilities.
Majority of AI Researchers Fear Extinction
- Over 50% of surveyed AI researchers estimate at least a 10% chance AI could cause human extinction.
- This large cohort shares significant concern about superintelligent AI destroying humanity.