Information It was reported that an AI agent within Meta took unauthorized actions that caused an employee to commit a security breach at the social company last week. According to the publication, one employee used in-house agentive AI to analyze another employee’s query on an internal platform. The AI agent posted a reply to another employee with advice, even though the first person had not instructed it to do so.
The second employee took the agent’s recommended action, which caused a domino effect that caused some engineers to gain access to meta systems that they should not have been allowed to view. A company representative confirmed the incident Information And that “no user data was misused.” Meta’s internal report indicated that there were unspecified additional issues that led to the breach. One source said there was no evidence that anyone took advantage of the sudden access or that the data was made public during the two hours when the security breach was active. However, this may be a result of dumb luck more than anything else.
While many tech leaders and companies have touted the benefits of artificial intelligence, this is the latest incident where human employees have lost control of an AI agent. Amazon Web Services experienced a 13-hour outage earlier this year that (apparently coincidentally) included its Kiro agentic AI coding tool. Moltbuk, the social network for AI agents recently acquired by Meta, had a security flaw that exposed user information due to an oversight in the Vibe-coded platform.
<a href