"Mythos Is Not Scary"...OpenAI Expands Cybersecurity Initiative

[Financial News] OpenAI said on the 23rd that it is expanding its AI-powered cybersecurity initiative, Daybreak, and has unveiled a new security model, GPT-5.5 Cyber.
Daybreak aims to speed up the entire remediation process, from finding software vulnerabilities to verification, risk assessment, patch development, testing and deployment.
As part of the Daybreak expansion, OpenAI also introduced the Codex Security plugin. Beyond simply generating alerts, Codex Security understands a team’s code and threat model to identify potential vulnerabilities. 
OpenAI said GPT-5.5 Cyber will be made available in a limited capacity to experts. The company explained that the model retains general intelligence and the ability to carry out long, complex tasks, while also helping discover and patch software vulnerabilities. 
OpenAI is also launching the Daybreak Cyber Partner Program, which will help selected security software and service partners use GPT-5.5 and a trust-based approach to cybersecurity in customer-facing products and services. Together with participating partners, the company plans to continuously strengthen the necessary safeguards, monitoring and anti-abuse standards.
OpenAI is also launching the 'Patch the Planet' program to help open-source project maintainers move from vulnerability discovery to real-world fixes. More than 30 open-source projects have expressed interest in joining the program. 
In preparation for the emergence of AI models with stronger cybersecurity capabilities, OpenAI is working closely with governments and institutions around the world, including the U.S. government, to strengthen defensive cybersecurity capabilities and protect critical infrastructure. 
Daybreak is expected to help defenders respond to cybersecurity threats by bringing together cutting-edge models and the surrounding ecosystem. 
Organizations in both the public and private sectors can work with OpenAI Daybreak to discover, verify and fix vulnerabilities in the software they develop or rely on. They can also use AI models to build safer software and environments with stronger cyber resilience.


wongood@fnnews.com Juwon Gyu Reporter