Prompt injection is role confusion, and your MCP gateway can't see it

There is a paper that reframes prompt injection in a way that is hard to unsee: Prompt Injection as Role Confusion. Its claim is that the dozens of named attacks (ignore previous instructions, hidden HTML, markdown injection, tool injection, RAG injection) are not different bugs. They are one bug: a

Related Stories

📰

Your Baby Monitor's Biggest Security Flaw Isn't Hackers. It's the Company That Built It.

TypeScript 5.7's `--module nodenext` Disrupts Legacy Express Applications

📰

OpenCode: a alternativa open source ao Claude Code (sem lock-in de modelo nem de nuvem)

📰

Sent 419 cold B2B emails: 41% opened, no clicks achieved.