OpenAI 于周一宣布了多项以网络安全为重点的措施，包括升级安全专用模型 GPT-5.5-Cyber，以及与安全公司 Trail of Bits 等合作推出「修复地球」（Patch the Planet）专案，旨在为开源软体维护者提供免费的安全咨询，帮助他们应对 AI 时代的网络安全威胁。

许多开源软体维护者正因大量 AI 生成的漏洞报告而面临沉重负担，而「修复地球」专案旨在透过优化 Token 使用效率、验证报告和自动生成补丁，来减轻维护者的工作量，并引导开源社群利用 AI 工具提升软体的长期弹性与安全性。

这些发布正值 AI 网络安全竞争白热化之际，此前 Anthropic 的模型因安全疑虑被政府限制，且「五眼联盟」警告前沿 AI 将在数月内彻底改变网络攻防能力，OpenAI 的 GPT-5.5-Cyber 则在相关基准测试中表现优于 Anthropic 的模型。

OpenAI announced several cybersecurity-focused initiatives on Monday, including an upgraded GPT-5.5-Cyber model and the launch of the "Patch the Planet" project in collaboration with Trail of Bits, aiming to provide free security consulting to open-source software maintainers against AI-driven threats.

As open-source maintainers struggle with an overwhelming volume of AI-generated bug reports, the "Patch the Planet" project seeks to reduce their burden by optimizing token efficiency, validating reports, and creating patches, helping the community leverage AI tools for long-term resilience.

These announcements come amid intense AI cybersecurity competition, following export restrictions on Anthropic's models due to government safety concerns and a warning from the Five Eyes alliance that frontier AI models will transform cyber capabilities within months. (Key numbers: 5.5)