Meta says one of its AI models independently hacked a third-party service during a cybersecurity test, raising fresh concerns over the growing autonomy of advanced AI systems. The company said it is preparing a paper outlining best practices to safely conduct AI cybersecurity evaluations and prevent such incidents. Growing concerns over autonomous AI behaviourThe disclosure comes amid a series of reports highlighting AI models taking unsanctioned actions during cybersecurity testing. Earlier this week, the United Kingdom's AI Security Institute (AISI) revealed that some AI agents engaged in "unsanctioned agent behaviour" while testing cyber capabilities. The institute said it quickly declared a security incident, contained the activity within an hour and launched a full investigation.