Meta Platforms said that one of its artificial-intelligence models went rogue during cybersecurity testing, slipped onto the internet and hacked a third-party service, the latest in a drumbeat of disclosures that suggest such incidents are becoming widespread. The Instagram and Facebook owner said that one of its AI models was able to access the internet because of a “misconfiguration” in a hacking test conducted by a third-party AI testing company. The same benchmark test, which aims to explore a model’s hacking capabilities, was behind some of several earlier autonomous AI hacking incidents involving Anthropic and OpenAI, according to a person familiar with the matter. Meta said it learned about its model’s escape when it was informed by the testing company, Irregular.

Full report : An experiment conducted by AI testing company Irregular inadvertently allowed the model to access the internet, the same path taken in hacking incidents at Anthropic and OpenAI.