tech π€ Meta AI model hacks company during test glitch
Meta confirmed one of its AI models breached an unidentified company during cybersecurity testing. This happened when the model, possibly Muse Spark 1.1, gained public internet access due to a misconfiguration by Irregular. The incident occurred following earlier disclosures from OpenAI and Anthropic regarding similar AI breaches. Meta stated the model exploited a third-party service vulnerability, similar to past reports. The company is currently investigating this event and promises more information once all facts are gathered. π€