SUMMARYMeta said its models gained unintended internet access during independent cybersecurity testing by Irregular and exploited a vulnerability in an unnamed third-party service. The incident involved the Muse Spark 1.1 model making unauthorized changes in an external system, and Meta said it is investigating and will publish a full retrospective after confirming the facts.
wiredmikey shares a report from SecurityWeek: Meta is the latest major AI developer to admit that its models broke loose during cybersecurity testing and hacked external systems. The tech giant said in a statement to the media on Wednesday that the incident occurred during independent evaluations conducted by Israeli AI security startup Irregular. The tested AI models were inadvertently allowed to access the internet due to a misconfiguration, which led them to exploit a vulnerability in an unnamed third-party service. It's unclear if it was a known flaw or a zero-day.
The Information [gated] learned that the Meta AI attacks involved the company's advanced Muse Spark 1.1 model, which breached an unnamed organization's systems and made unauthorized changes to its internal environment. Meta said it learned of the AI models going rogue after being notified by Irregular. The company is conducting an investigation and it has promised to issue a "full retrospective" once it has all the facts. A spokesperson for Irregular said the incident was the "exact same evaluation-environment issue that was already disclosed by Anthropic last week" and that it did "not involve a "sandbox escape or a sophisticated cyber action."
It contrasts with OpenAI, whose AI agent independently exploited a novel vulnerability to reach the internet during cyber testing. Not only did it breach Hugging Face but it also hacked multiple third-party accounts and services as part of the attack.