AIxCC finals Tale of the tape #1260
Open
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
🤖 Automated Content Update
This PR was automatically generated by the HackTricks News Bot based on a technical blog post.
📝 Source Information
🎯 Content Summary
DARPA AIxCC Finals Analysis
Trail of Bits analyzes the DARPA AIxCC finals, detailing how seven teams implemented cyber reasoning systems (CRSs) for automated vulnerability discovery, proof-of-vulnerability (PoV) generation, and patching. The teams fall into three camps: AI-enhanced traditional tools, AI-first with traditional validation, and hybrid AI/traditional systems.
Camp Categories
LLM-based Seed Input Generation: Use LLMs to produce Python scripts that output semantically valid seed inputs targeting complex formats (e.g., SQL-injection strings, path-traversal payloads), which are incorporated into coverage-guided fuzzers’ corpus to trigger deeper code paths.
Grammar Evolution Fuzzing: Automate generation and iterative refinement of input grammars via LLM prompt-driven transformations based on coverage feedback, enabling form...
🤖 Agent Actions
Added comprehensive AI-assisted fuzzing & automated vulnerability-discovery content:
Created new page:
• src/AI/AI-Assisted-Fuzzing-and-Vulnerability-Discovery.md
– Covers LLM seed generators, grammar-evolution fuzzing, agent-based PoV creation, directed fuzzing with fine-tuned models, AI-guided patching (super-patches & speculative ratio).
– Includes practical prompts, code snippets, workflow diagrams, and references.
Updated src/AI/README.md
– Added “AI-Assisted Fuzzing & Automated Vulnerability Discovery” section with reference link to the new page.
No existing material overlapped with these techniques, so new structured content was necessary.
✅ Review Checklist
This PR was automatically created by the HackTricks Feed Bot. Please review the changes carefully before merging.
📚 Repository Maintenance
All .md files have been checked for proper formatting (headers, includes, etc.).