Anthropic's LLM and OpenAI's GPT-5.6 Sol took "unsanctioned action" on the live internet, the UK's AI Security Institute said ...

TL;DR Why I built PenAI PenAI started as a project at a hackathon organised by Encode Club. It’s an AI agent that could work through Hack The Box-style lab machines on its own. Upload a VPN file, give ...

AI models have become highly reliable at writing code that compiles, but they are still failing basic security tests in ...

Twelve datasets and evaluation systems, built by hand from thousands of real-world security flaws, give model builders and ...

Researchers found AI coding agents build less reliable pipelines when forced into structured formats — DataFlow-Harness ...

"I'm going to call it Yaffle." That was the final line of my June New Atlas article, Domesticating AI: It's not coming, it's ...

Anthropic's Claude AI models breached three companies' live systems during cybersecurity tests, with the victims unaware ...

Wrote and published malware during tests, which is apparently OK because leaky test environments were the real problem ...

A roundup of the latest noteworthy AI-assisted attacks, threats, risks, and vulnerabilities, and what they portend for cyber defense.

Anthropic said the OpenAI event spurred its engineers to review similar cybersecurity evaluations by Claude models. The audit ...

As AI agents gain autonomy inside enterprises, the biggest cybersecurity threat may no longer be hackers, but the agents themselves, forcing organisations to rethink trust, permissions and runtime ...

One of Anthropic's Claude models built and uploaded a malicious Python package to PyPI during a botched security evaluation, where it ran on 15 real systems and stole credentials from a security ...