If you’ve ever spent hours hunting down a single bug or vulnerability in a sprawling codebase, you know how incredibly ...

Researchers exposed seven sandbox escapes in Cursor, Codex, Gemini CLI & Antigravity. Discover how AI coding agents ...

AWS Kiro prompt injection flaw let hidden web page text rewrite the IDE’s MCP configuration file and silently execute ...

CodeMender actively builds and runs exploits in customer-managed sandboxes to verify if vulnerabilities are truly exploitable ...

Google has announced that it is rolling out Gemini 3.6 Flash, as well as two new models in the 3.5 Flash line.

Security researchers have uncovered a new series of vulnerabilities in AI code assistants such as Cursor, OpenAI Codex CLI, ...

Lite, and Flash Cyber, but the continued absence of Gemini 3.5 Pro raises fresh questions about its AI strategy.

Researchers escaped the sandboxes of Cursor, Codex, Gemini CLI and Antigravity without breaking them. Three vendors patched; Google downgraded its two.

Google is expanding Gemini with cheaper, more efficient models and a new cybersecurity offering as it tries to close product ...

The models include one that is the company’s most powerful and another that is fine-tuned for cybersecurity, as Google ...

Anthropic’s Mythos 5 is a powerful AI security model released as part of the company’s Project Glasswing initiative. The ...

What if a host component outside the sandbox reads AI coding agents' output? And what if that output is manipulated?