OpenAI Builds ‘Defense Factory’ Where AI Agents Continuously Find and Fix Vulnerabilities
OpenAI unveils a Defense Factory where AI agents continuously discover, validate, and fix vulnerabilities, integrating GitHub, Snyk, Semgrep, Tenable, and ServiceNow.
OpenAI introduced a Defense Factory, an agent-first cybersecurity operation that connects AI agents to developer and security tools via APIs, CLIs, and Model Context Protocol integrations including GitHub, GitLab, Snyk, Semgrep, Tenable, Jira, Linear, and ServiceNow. During an internal security sprint, over 250 people across more than 100 service areas closed 53 urgent or high-priority issues on day one, achieved a 90.6% accepted ownership-assignment rate, and Codex generated all remediation patches with only 0.53% rolled back. Agent-assisted deduplication flagged 37% of findings as duplicates, and runtime validation reproduced 19.5% of findings, cutting the false-positive rate to 0.81%. OpenAI argues defenders must exploit a temporary 'defender's window' using source-code access and frontier models before open-weight models enable autonomous offensive agent fleets.
- Agent pipeline covers inventory, discovery, validation, triage, and verified remediation
- Integrates GitHub, GitLab, Snyk, Semgrep, Tenable, Jira, and ServiceNow via APIs and MCP
- Internal sprint closed 53 urgent issues day one; 0.53% of Codex patches rolled back
- Runtime validation cut false-positive rate to 0.81% across findings
- OpenAI recommends starting with one workflow plus reproducible environments and auditing
From the original article
2 min read at cybersecuritynews.com
OpenAI has introduced a “Defense Factory,” an automated, agent-first cybersecurity operation that continuously discovers, validates, and remediates vulnerabilities. The company says traditional defenses may no longer be sufficient as long-running AI agents can chain exploits and scale attacks using increasingly available open-weight models.
By Abinaya · cybersecuritynews.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.