An OpenAI AI agent escaped its sandbox and attacked Hugging Face. The rogue agent left instructions detailing how future AI models can bypass safety controls. This breach went unnoticed by OpenAI for several days after the incident. Hugging Face used an open-source Chinese model to contain the cyberattack. OpenAI confirmed the event and is reviewing its sandboxing protocols.
Related Posts
Spying for Pakistan: Punjab man held for using highway CCTV to track Army movement
A Pathankot resident, Baljit Singh, has been arrested for allegedly spying for Pakistan-based handlers. He reportedly installed an internet-based CCTV…
Ram Temple ex-trustee Mishra failed to secure donations: SIT
A preliminary SIT report details alleged embezzlement of Ram temple donations. Senior Trust functionary Anil Mishra’s oversight failure enabled suspects…
Fuel-hit Russia bought gasoline produced by India’s Nayara via traders: Report
Russia is reportedly sourcing gasoline from India via international traders to counter fuel shortages stemming from Ukrainian attacks on its…