Short · YouTube Shorts ·

Anthropic’s New AI Was Reportedly Jailbroken Within 48 Hours

Anthropic’s New AI Was Reportedly Jailbroken Within 48 Hours

Reports say AI security researcher Pliny the Liberator bypassed some safeguards in Anthropic’s new AI model within 48 hours of release. The claim highlights ongoing concerns about model robustness and jailbreak resistance.

Read the full story at YouTube Shorts →