OpenAI says its models found critical PHP flaws affecting ~70% of websites
OpenAI · x · 2026-10-22
OpenAI announced that models used in its Patch the Planet initiative helped researchers discover critical security flaws in PHP, which powers an estimated 70% of websites. Over two months, OpenAI coordinated maintainers, researchers, and security vendors on fixes requiring major changes to the PHP engine; sites are safe once they upgrade PHP and apply vendor mitigations.
To help admins self-check, OpenAI open-sourced the php2shell-checker repo with Codex skills: phpcheck-server for those with server access, phpcheck-url for URL-only checks, or node phpcheck.js directly — returning OK / VULNERABLE / UNKNOWN verdicts.
Related event: OpenAI Models Uncover Critical PHP Flaw Affecting 70% of Websites(48 posts)→
More from coding & agent
- Mastercard teams with Coinbase on x402 agentic payments for AI agents — MurrLincoln · 2026-10-22
- Recursive's Flint: 39.8B MoE local coding model hits 125+ tok/s on a MacBook Pro — ivan_bezdomny · 2026-10-22
- Recursive open-sources Flint, a 39.8B MoE coding-agent model with 2.5B active params — ollama · 2026-10-22
- Harvey builds a frontier marketing org on Figma MCP, with agents running design projects end to end — saranormous · 2026-10-22
- AgentMail offers email inbox API for AI agents, now handling 1B+ requests a month — mhdfaran · 2026-10-22
- HarnessAdapt-RL trains the coding agent's harness too, beating GRPO by 9.8 points on SWE-bench Verified — tencent · 2026-10-22