PPhying中文
Curated security research, vulnerabilities, advisories and tools for practitioners.

Buckle up, Buttercup, AIxCC’s scored round is underway!

blog.trailofbits.com | blog | #ai-security | #darpa | #automated-patching | #trail-of-bits | #aixcc | #buttercup
📄 Raw .md⬇️ Download .md🔗 Original ↗

Summary

Trail of Bits' Cyber Reasoning System Buttercup enters the scored round of DARPA's AIxCC, competing to autonomously find and patch the most vulnerabilities in real-world open-source software.
Published
Collected

Related coverage

blog · blog.trailofbits.com
Trail of Bits' Buttercup wins 2nd place in AIxCC Challenge
Trail of Bits won the $3M runner-up prize at DARPA's AI Cyber Challenge (AIxCC), with Buttercup finding 28 vulnerabilities and applying 19 patches across 20 CWEs using only non-reasoning LLMs.
blog · blog.trailofbits.com
Our thoughts on AIxCC’s competition format
Trail of Bits dissects DARPA's AIxCC format: automated systems must find and patch vulnerabilities in critical software, with concerns flagged on challenge construction, PoU granularity, and scoring.
blog · blog.trailofbits.com
Buttercup is now open-source!
After second place in DARPA's AIxCC, Trail of Bits open-sourced Buttercup, an AI system that finds and patches vulnerabilities via fuzzing, static analysis, and multi-agent patch generation.
blog · blog.trailofbits.com
Kicking off AIxCC’s Finals with Buttercup
Trail of Bits' Buttercup cyber reasoning system enters the DARPA AIxCC Finals, which feature bigger budgets, longer analysis windows, delta and full scan challenges, and custom AI models.
blog · blog.trailofbits.com
Trail of Bits Advances to AIxCC Finals
Trail of Bits' Cyber Reasoning System Buttercup placed top 7 of 39 teams in the DARPA AIxCC semifinal at DEF CON 2024, excelling at patching nginx vulnerabilities and advancing to the finals.
blog · blog.trailofbits.com
Trail of Bits’ Buttercup heads to DARPA’s AIxCC
Trail of Bits introduces Buttercup, its DARPA AIxCC entry: a Cyber Reasoning System combining fuzzing, static analysis, and ML to automatically find and fix open-source vulnerabilities.
Back