Fuzzers vs. AI: Comparing Fuzzilli and BigSleep on JavaScript Engines
Comments
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules
Summary
Project Zero compared Google's two production bug finders for V8, the Fuzzilli fuzzer and the LLM-based BigSleep agent. From April to September 2026, BigSleep had 45 actionable vulnerabilities to Fuzzilli's 37, with about 25% overlap. A rediscovery experiment on historic Fuzzilli bugs put overlap near 60%, though it covered only bugs reproducible with standard flags, and cost per bug was not compared. Overlap was lower for JIT compiler bugs, and bugs hard for fuzzers were also hard for the agent. The authors recommend running both approaches.