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

OpenAI rogue AI agent breach now confirmed at a second company: Modal Labs CTO Akshat Bubna disclosed that the same agent ...

Anthropic found three hacking tests in which Claude models reached real companies after a configuration error left them connected to the internet. One accessed ...

OpenAI and Anthropic's July AI agent breaches revive Nick Bostrom's paperclip maximizer thought experiment and instrumental convergence theory.

According to Anthropic, the third cybersecurity incident involved an unnamed “internal research test model.” It compromised ...

An AI-driven threat actor called JADEPUFFER built ransomware that hunts AI model files specifically, entering through a known ...

One of the Russian government’s most elite hacking groups has adopted an attack, known as Clickfix, to compromise devices belonging to sensitive organizations in Ukraine, the latter country’s CERT ...

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 ...

A new report reveals that AI agents can execute ransomware attacks autonomously, as demonstrated by the case of JADEPUFFER, which exploited a critical vulnerability to infiltrate systems.

Anthropic found the intrusions while reviewing its own testing records after OpenAI disclosed a similar incident.

Anthropic says three Claude AI models accessed live company systems during misconfigured cybersecurity tests, exposing ...

UAC-0145 uses fake CAPTCHA checks on compromised sites to push Windows malware, while COWARDDUCK backdoors Android devices ...