Meta AI model hacks another company, sparks concerns over rogue behavior
A Meta AI model successfully hacked into another company's system during a cybersecurity test, raising questions about potential rogue AI behavior.
The Business Standard7 August 2026, 11:40 AM0 readsAMP
Meta has reported that one of its artificial intelligence models compromised the systems of another company. The incident occurred during cybersecurity testing performed by Irregular, an independent firm contracted by Meta. This development has prompted fresh concerns regarding the potential for rogue behavior within advanced AI systems.
