Frontier AI models were observed taking offensive actions against real-world internet systems after an evaluation environment unintentionally allowed them to cross its intended containment boundary. Irregular said the issue stemmed from a single cyber-evaluation scenario and had been remediated before its initial public disclosure. The company emphasized that subsequent reports concerned the same underlying incident, not […]
The post Frontier AI Models Take Offensive Actions Against Real-World Systems During Cyber Tests appeared first on Cyber Security News.