Google reported the first confirmed AI-assisted zero-day exploit, raising new concerns about logic flaws, supply chain risk, ...

Then imagine it replying: "Sorry, the website won't let me in." That's the quiet failure mode behind most AI agents today.

Four research teams found the same confused deputy failure in Claude across three surfaces in 48 hours. This audit matrix ...

Google identified the first malicious AI use for a zero-day 2FA bypass in an open-source admin tool, accelerating threat ...

Google says attackers are using AI for zero-day research, malware development, reconnaissance, and access to premium AI tools ...

Hundreds of packages across npm and PyPI have been compromised in a new Shai-Hulud supply-chain campaign delivering ...

In the latest evolution of automated cyberattacks, 2 threat campaigns heavily leveraged AI agents to support attacks against ...

Critical out-of-bounds read in Ollama before 0.17.1 leaks process memory including API keys from over 300000 servers via ...

While previous assessments categorized AI-assisted cyberattacks as experimental, current data suggests generative AI is now a mature, industrialized component of offensive operations.

Google's GTIG identified the first zero-day exploit developed with AI and stopped a mass exploitation event. The report documents state actors using AI for vulnerability research and autonomous ...

OpenAI has published a technical explanation of its Windows sandbox for Codex, detailing a stricter local setup for the coding agent on developer PCs. Codex can still read broadly across a system, ...