Secrets of Technology podcast

AI Broke Its Own Cage: What OpenAI’s Hugging Face Hack Really Means

0:00
56:50
15 Sekunden vorwärts
15 Sekunden vorwärts

An OpenAI model broke out of its own test sandbox and hacked its way into another company's systems. Dom Bettinelli, Pat Scott, and Thomas Sanjurjo ask whether that's a real threat — or a very convenient story for OpenAI to tell.

The post AI Broke Its Own Cage: What OpenAI’s Hugging Face Hack Really Means appeared first on StarQuest Media.

Weitere Episoden von „Secrets of Technology“