Researchers bypass GPT-5 guardrails using narrative jailbreaks, exposing AI agents to zero-click data theft risks.

A vulnerability that researchers call CurXecute is present in almost all versions of the AI-powered code editor Cursor, and ...

AI agents: the new insider riskThe biggest internal security threat may now come from employees — unintentionally.According to a new survey of 200 North American security leaders, 63% say the top ...

Cybercriminals are abusing the link wrapping security feature from Proofpoint and Intermedia to launch sophisticated phishing ...

Critical flaw in Cursor AI editor let attackers execute remote code via Slack and GitHub—fixed in v1.3 update.

Anthropic has released a new safety framework for AI agents, a direct response to a wave of industry failures from Google, ...