Partnering with Mozilla to improve Firefox’s security

Anthropic and Mozilla teamed up to validate AI-discovered Firefox security flaws.

Using Claude Opus 4.6, Anthropic reproduced and found 22 vulnerabilities in two weeks, including 14 high-severity bugs.

Mozilla confirmed the findings and shipped fixes in Firefox 148.0, including issues such as use-after-free bugs.

The collaboration suggests AI can accelerate security research and uncover serious browser vulnerabilities much faster than traditional methods.