GitHub - yashmahajan10/llm-differential-privacy-gateway: Noisegate: a differential privacy gateway that lets an untrusted LLM agent query sensitive data over MCP (Model Context Protocol), with a formal guarantee no individual's record can leak even if the agent is adversarial - enforcement lives in trusted code below the model, validated by a runnable attack gallery.

Show HN: Noisegate – a differential-privacy gateway for untrusted AI agents

Noisegate: a differential privacy gateway that lets an untrusted LLM agent query sensitive data over MCP (Model Context Protocol), with a formal guarantee no individual's record can leak even i...

19 points by yashmahajan10 84 days ago github.com 0 comments

Summary

Noisegate offers a solution for safely querying sensitive data using untrusted AI agents while ensuring individual privacy. It employs differential privacy techniques to prevent any risk of personal data exposure.

Comments (0 of ) ⌛

Loading more comments... ⌛

Scroll down to load more comments

No comments available.