Anthropic's Opus 4.6 system card breaks out prompt injection attack success rates by surface, attempt count, and safeguard ...

These 4 critical AI vulnerabilities are being exploited faster than defenders can respond ...

Three security vulnerabilities in the official Git server for Anthropic's Model Context Protocol (MCP), mcp-server-git, have been identified by cybersecurity researchers. The flaws can be exploited ...

"Ever wonder what an AI’s ultimate high looks like?" The post Bots on Moltbook Are Selling Each Prompt Injection “Drugs” to ...

Microsoft has pushed back against claims that multiple prompt injection and sandbox-related issues raised by a security engineer in its Copilot AI assistant constitute security vulnerabilities. The ...

Attackers are doubling down on malicious browser extensions as their method of choice, stealing data, intercepting cookies and tokens, logging keystrokes, and more. Join Push Security for a teardown ...

OpenAI launches Lockdown Mode and Elevated Risk labels in ChatGPT to curb prompt injection threats and protect sensitive user data.

Some of the latest, best features of ChatGPT can be twisted to make indirect prompt injection (IPI) attacks more severe than they ever were before. That's according to researchers from Radware, who ...

A single prompt can now unlock dangerous outputs from every major AI model—exposing a universal flaw in the foundations of LLM safety. For years, generative AI vendors have reassured the public and ...

As OpenAI and other tech companies keep working towards developing agentic AI, they’re now facing some new challenges, like how to stop AI agents from falling for scams. OpenAI said on Monday that ...