CaMeL: Defeating Prompt Injections by Design (arxiv.org)
recommended by 2 users | 2025 | 71 points by tomrod | ai and society | original thread |
2 users recommended this thread. What they said:
M4v3R May 26, 2025 | on: Remote Prompt Injection in Gitlab Duo Leads to Source Code… | context

DeepMind recently did some great work in this area: https://news.ycombinator.com/item?id=43733683

The method they presented, if implemented correctly, apparently can effectively stop most prompt injection vectors

crazygringo Sep 26, 2025 | on: How to stop AI's "lethal trifecta" | context

Thank you! That is very helpful.

I'm very surprised I haven't come across it on HN before. Seems like CaMeL ought to be a front-page story here... seems like the paper got 16 comments 5 months ago, which isn't much:

https://news.ycombinator.com/item?id=43733683

About | Topics | Past | Comments | Recent | Hindsight | Data | Keyboard shortcuts
Hacker News for Hacker News
Not affiliated with Y Combinator or Hacker News. Comments are by their HN authors. Data: open-index/hacker-news (ODC-BY) and the HN API, as of November 3, 2026.