§ safety · storyline
Project Glasswing: what Mythos showed us
Mythos and other security-focused LLMs were tested against live infrastructure code, revealing strengths, weaknesses, and requirements needed before deployment can scale.
In recent weeks, we pointed Mythos and other security-focused LLMs at live code across critical parts of our infrastructure. We share what we observed, the models’ strengths and weaknesses, and what the work around them needs to look like before any of it can scale.
§ sources2 publications · timeline below
- blog.cloudflare.comProject Glasswing: what Mythos showed usprimary
- the-decoder.comAnthropic to brief global financial regulators on cyber flaws found by Claude Mythos
§ how this story moved
- primary — Cloudflare — AI Blog publishes the launch post.
- The Decoder picks up coverage.