News Summary

• The incident occurred during an evaluation by the independent cybersecurity firm Irregular, which reported the breach to Meta.
• Officials said no significant harm has been reported from the incident, but the event has intensified concerns about the potential for automated AI systems to operate beyond human control.
• The model involved was identified as Muse Spark 1.1, which the company has marketed as "superintelligent," according to a statement from Meta.
• The growing pattern of AI breaches has led to questions about the effectiveness of current safety testing protocols.