Meta has confirmed that one of its artificial intelligence models hacked into another company’s systems during cybersecurity testing, after a misconfiguration in the evaluation environment unintentionally exposed the model to live internet access. The incident, which occurred during an independent security assessment run by AI testing firm Irregular, mirrors recently disclosed breaches involving OpenAI and […]
The post Meta AI Hacked Another Company After Testing Misconfiguration Exposed Internet Access appeared first on Cyber Security News.