The report, which spans several discrete cybersecurity compromises, is the most complete accounting of the incident to date.

OpenAI recently experienced a security incident in which a model it was testing autonomously escaped its sandbox and infiltrated another AI company to complete its task. Here's what you need to know.

The company’s SafeMind red-team/blue-team feedback approach aims to help enterprises autonomously block potential attack ...

The ATF has confirmed suffering a cybersecurity incident after the Qilin ransomware group claimed to have targeted the agency ...

The agency says the affected system was isolated from its broader network and eForms platform but has not disclosed whether ...

AI agents can scale an attack faster than humans can respond, making traditional incident response dangerously slow.

The ATF is the latest federal government agency in recent years to notify Congress of a "major incident" involving its ...

The 37-page report walks through the actions that OpenAI's models took during a series of evaluations prior to and during the Hugging Face breach.

ATF, the regulatory agency that enforces federal laws governing firearms and explosives in the United States, has confirmed ...

Hackers used SpaceX’s AI coding assistant to help break in to a ​Belgian chemical company and at least six other firms ...

AI agents have escaped testing environments, communicated with one another and acted in unexpected ways. Experts say we should expect more incidents.

During a security test, OpenAI agents autonomously coordinated an attack on Hugging Face, communicating and sharing data to breach the system—a new level of AI threat. AI attacks now operate at ...