**[OpenAI Releases Full Version of GPT-5.5-Cyber, CyberGym Scores Surpass Mythos 5]** Odaily Planet Daily News: OpenAI has announced an upgrade to its Daybreak network defense toolchain and officially launched the full version of the GPT-5.5-Cyber model tailored for cybersecurity defense scenarios. In the CyberGym benchmark test, which evaluates AI agents' ability to replicate known vulnerabilities, GPT-5.5-Cyber achieved a single-model score of 85.6%, surpassing GPT-5.5's 81.8% and Anthropic Mythos 5's 83.8%. OpenAI stated that as AI significantly improves vulnerability discovery efficiency, the core bottleneck in the cybersecurity field is shifting from "discovering vulnerabilities" to "automatically patching vulnerabilities." To address this, the company has simultaneously upgraded the Codex Security plugin, enabling developers to automatically analyze vulnerabilities and generate patches within the Codex environment. Since the release of the preview version in March this year, Codex Security has scanned over 30 million code submissions and automatically confirmed fixes for 500,000 security flaws. The new version also supports CodeQL query integration and SARIF standard file export. Additionally, OpenAI has partnered with organizations such as Trail of Bits and HackerOne to launch the "Patch the Planet" open-source security project, providing ChatGPT Pro subscriptions and API credits to over 30 mainstream open-source projects, including cURL and Go. A team of security experts manually verifies the quality of the patches. Regarding the Daybreak Partner Program, security vendors such as Palo Alto Networks and Wiz have already integrated these capabilities.