Saving Bug Bounties from AI Slop
summary
Bug bounty platforms drown in AI-generated spam and nonreproducible reports, so the talk demonstrates a working system that uses ZK-TLS/TLSNotary MPC attestations to produce cryptographic proofs of HTTP request/response pairs, a Burp Suite extension and Docker notary that generate redacted proofs, and an optional redacted request reveal flow that ties a later plaintext request to the original proof and unlocks payment. Demo shows end-to-end reporting, automated verification, and a working HackerOne-style flow but notes MPC latency and ongoing API instability.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:04Bug bounties reality
- 3:36Zero knowledge basics
- 12:56ZK-TLS explained
- 16:49Trustless bug bounties
- 18:57Live demo walkthrough
- 27:24Next steps and research
- 28:26Q&A session
related talks
53mOpening Keynote – Dr. Katie Paxton-Fear
42mHidden Chains: Revealing High-Impact Bugs from Bounty submissions
41mThe Future Is Trustless: How ZkVMS And ZkTLS Transform Bug
46mHow to run a successful bug bounty programme
20mBillion Dollar Bugs
29mAPI Security Risk Management with Bug Bounties
23mThanks for all the bugs!
50mBug Bounty Programs: Successfully Controlling Complexity and Perpetual Temptation