* Axios article
Mozilla fixes 22 security flaws flagged by Anthropic's AI
https://www.axios.com/2026/03/06/anthropic-mozilla-claude-opus-bug-hunting
Summary of part of the article.
Mozilla announced last Friday that it had fixed over 100 Firefox bugs, including 22 security vulnerabilities, found by Anthropic's Claude.
https://blog.mozilla.org/en/firefox/hardening-firefox-anthropic-red-team/
.
AI models are significantly lowering the cost of finding software vulnerabilities and uncovering substantial flaws even in rigorously tested projects like Firefox.
Anthropic discovered over 500 previously unknown bugs across open-source projects while testing Claude Opus 4.6 last month, including 112 submitted to Mozilla over two weeks.
https://www.anthropic.com/news/mozilla-firefox-security
.
Anthropic also launched an automated code security testing tool called 'Claude Code Security' last month, which briefly caused fluctuations in cybersecurity stocks.
[Rest of the content omitted]
(February 2026 Claude Opus 4.6 found 22 security vulnerabilities.)
