Behind the Scenes Hardening Firefox with Claude Mythos Preview
Simon Willison · Simon Willison · 2026-05-07
(No summary yet for this item — extraction summaries are still backfilling.)
Appears in
Extraction
Topics: ai-security-researchclaude-mythosvulnerability-detectionllm-applicationsfirefox
Claims
- Mozilla used early access to Claude Mythos to find and fix hundreds of security vulnerabilities in Firefox.
- Firefox security bug fixes jumped from 20-30 per month throughout 2025 to 423 in a single month (April) using AI-assisted analysis.
- The improvement in AI-generated security reports came from both more capable models and significantly better orchestration and filtering techniques.
- Many AI-discovered exploit attempts were stopped by Firefox's existing defense-in-depth measures, validating that layered security still holds.
Key quotes
It is difficult to overstate how much this dynamic changed for us over a few short months. This was due to a combination of two main factors. First, the models got a lot more capable. Second, we dramatically improved our techniques for harnessing these models — steering them, scaling them, and stacking them to generate large amounts of signal and filter out the noise.
Mozilla were fixing around 20-30 security bugs in Firefox per month through 2025. That jumped to 423 in April.