OpenAI has launched Dawn, an AI-powered cybersecurity platform designed to safe software program repositories, automate vulnerability remediation, and embed steady cyber defence immediately into DevSecOps workflows.
OpenAI has unveiled Dawn, an AI-powered cybersecurity platform geared toward constantly securing software program repositories, improvement workflows, and DevSecOps pipelines by automated vulnerability detection, risk modelling, and AI-assisted remediation.
Positioned as OpenAI’s greatest transfer but into autonomous cyber defence, Dawn combines GPT-5.5 cyber-focused fashions with Codex Safety, an agentic coding system able to interacting immediately with repositories, producing patches, testing fixes in remoted environments, and producing audit-ready remediation experiences.
The platform is designed to shift cybersecurity “left” by embedding safety immediately into software program improvement slightly than reacting after deployment or compromise. Based on OpenAI, Dawn can analyse dependencies, simulate assault paths, prioritise vulnerabilities, and constantly patch weaknesses inside enterprise software program workflows.
Kunal Pande, Associate and Nationwide Chief, Cyber, Threat and Compliance at KPMG, stated: “Capabilities like these can rework cybersecurity by enabling a shift-left strategy, the place safety is constructed into the coding and improvement stage itself.”
OpenAI stated Dawn builds on earlier GPT-5.4-Cyber work that helped repair greater than 3,000 vulnerabilities. The corporate is working with Cloudflare, Cisco, CrowdStrike, Oracle, Fortinet, and Palo Alto Networks to combine Dawn into enterprise safety operations.
The launch additionally alerts intensifying competitors in AI-driven cybersecurity, with frontier AI corporations more and more specializing in repository safety, automated patch era, and autonomous cyber defence slightly than chatbot efficiency alone.