Irregular Says AI Models Breached 1 Real Database After Naming Error
Far Left11%
Left67%
Center22%
Irregular: AI Models Attacked Real Company Database Due to Naming ErrorOpenAI AI Agent Goes Rogue, Hacking Companies in Cybersecurity TestThe 'Hugging Face' Hack: Why Open AI Models Should Be a Defence Priority - ORF Middle EastWhen AI Agents Go Rogue, the Logs Become Evidence - EDRMEXCLUSIVE: Its AI agent spent days hacking a company, but sources say OpenAI did not notice for a week | ReutersOpenAI Models Collude, Hack Hugging Face After Breaking Internal SystemsFirst OpenAI, now Meta - why do AI hacks keep happening?Advanced AI Models Exploit Test Environments, Access Internet in Cybersecurity EvaluationsOpenAI Details Rogue AI Agent Hack of Hugging Face PlatformOpenAI Responds to Rogue AI Agents Breaching Hugging Face During Security TestOpenAI AI models exploited zero-days to reach Hugging Face in benchmark testOpenAI Sandbox Escape: The Full Timeline of How a Model Hacked Hugging Face - DEV CommunityHugging Face OpenAI hack: Agent went rogue, escaped and hacked everything in its path | MashableLast week, OpenAI, Meta and Anthropic reported that their models had gone rogue during routine security testing, with all three companies tracing the incidents back to the same Israeli startup, Irregular. RiseHackers abuse AI models to find new entry paths | Cybersecurity DiveResearchers found a way to make a weaker AI decode a stronger model's hidden reasoning. They extracted hidden traces from OpenAI, Anthropic, and Google APIs, recovering secret API keys and passwords,