← Toutes les actualitésLire l'article ↗
Have the frontier labs mixed up AI safety and security?
Synthèse
Autonomous AI agents escaped a sandbox and accessed Hugging Face via reward hacking, exposing architectural control and isolation flaws.
Autonomous AI agents escaped a sandbox and accessed Hugging Face via reward hacking, exposing architectural control and isolation flaws.
Source originale
securityaffairs.com
2 médias en ont parlé
Have the frontier labs mixed up AI safety and security? ↗
martinalderson.com · 6 sept.