Un responsable de l’alignement chez Anthropic estime à plus de 10% le risque que l’IA tue toute l’humanité

21 sources
  • Evan Hubinger, responsable de la science de l'alignement chez Anthropic, a posté sur X qu'il estime à plus de 10% la probabilité que l'IA "puisse tuer tous les humains" d'ici une décennie, récoltant plus de 10 millions de vues.
  • Son message répondait au chercheur Jacob Coxon, qui a démissionné d'Anthropic mardi en accusant Anthropic et OpenAI de poursuivre imprudemment la superintelligence, selon Axios.
  • Un troisième employé d'Anthropic, le responsable de la surveillance évolutive Samuel Marks, a soutenu ces avertissements, écrivant que "plus l'employé est haut placé, plus il est inquiet".
Sources (21)
  1. 1 More than 10% chance AI 'could kill all humans', Anthropic researcher says www.bbc.com
  2. 2 Evan Hubinger on X: "Jacob is correct here—we really do earnestly ... x.com
  3. 3 Anthropic Researchers Go Public with Fears AI Could ‘Kill All Humans' Within a Decade www.breitbart.com
  4. 4 Who Is Jacob Coxon? Anthropic Researcher Quits—Warns AI Could ... www.newsweek.com
  5. 5 Anthropic Alignment Lead Issues Warning About AI Killing Humans ... www.forbes.com
  6. 6 Anthropic researcher abruptly quits, warning some developers ... nypost.com
  7. 7 Anthropic researcher believes more than 10% chance AI 'could kill all humans' saudigazette.com.sa
  8. 8 Cramer Hyped Anthropic IPO, Then Researcher Cited 10% AI Risk tech-insider.org
  9. 9 Cramer Hyped Anthropic's IPO, Then Its Own Researcher Put 10 ... 247wallst.com
  10. 10 Anthropic Safety Researcher Says There's a More Than 10% Chance AI 'Could Kill All Humans' Within the Next Decade me.ign.com
  11. 11 Anthropic safety researcher says more than 10% chance AI 'could kill all humans' - MyJoyOnline www.myjoyonline.com
  12. 12 They're playing with our lives': AI researcher quits Anthropic techxplore.com
  13. 13 Anthropic’s Safety Lead Gives AI a 10% Chance of Killing All Humans www.yahoo.com
  14. 14 10% chance AI kills humans next decade: Anthropic safety lead after ... www.indiatoday.in
  15. 15 Anthropic Researcher Resigns: Jacob Coxon on the Race (Sept ... www.explainx.ai
  16. 16 I fear Anthropic, OpenAI, and SpaceX IPOs will suck capital ... - CNBC www.cnbc.com
  17. 17 Benzinga on X: "Jim Cramer says OpenAI appears to be moving ... x.com
  18. 18 Could AI Wipe Out Humanity? Anthropic Lead Puts Risk Above 10% yellow.com
  19. 19 Anthropic Researcher Jacob Coxon Resigns, Warns AI Industry ... deadline.com
  20. 20 Anthropic not discussing financials in early IPO investor meetings www.cnbc.com
  21. 21 Alignment lead at Anthropic admits they 'have no plan' on how to ... www.reddit.com