IBM announced on June 22 that it has joined OpenAI's Daybreak Cyber Partner Program, rolling out a new application security service built on OpenAI's frontier cyber capabilities. The service goes beyond traditional code scanning to identify and validate vulnerabilities at machine speed, framed by IBM as a necessary response to attackers who are already using AI to probe and exploit systems faster than defenders can keep up.

IBM said it has joined the OpenAI Daybreak Cyber Partner Program, bringing advanced frontier AI capabilities into enterprise security operations to help organizations counter machine-speed threats. The move builds on IBM's previously announced Project Lightwell and marks a concrete product launch rather than just a partnership announcement.

What's New

The new application security service moves beyond conventional static code scanning. It uses OpenAI's cyber capabilities to analyze application code and prioritize the areas most likely to contain flaws and exploitable paths. The underlying security harness runs on IBM Consulting Advantage, IBM's AI platform for delivering consulting services, and connects client application environments to advanced AI in a controlled, secured, and governed way.

Critically, the system operates with read-only access to code repositories and bounded execution — a design choice aimed at enabling large-scale exposure analysis without expanding the attack surface or giving the AI system write access to production code. IBM is positioning this as a managed enterprise service: clients can start with focused evaluations of key applications and later expand to continuous monitoring that reassesses risk as code changes and new threats emerge.

$5 billion Combined IBM and Red Hat investment in Project Lightwell
Available today The new application security service is live now
Read-only Level of code repository access, central to the security model

Connecting to Project Lightwell

The announcement ties directly into Project Lightwell, the $5 billion initiative backed by IBM and Red Hat that combines an enterprise security clearinghouse with a global engineering workforce to patch, validate, and manage open-source code across the software supply chain. Through this partnership, OpenAI's cyber capabilities — alongside other frontier AI models — will be applied to code review and vulnerability remediation at scale.

Category Detail
Program OpenAI Daybreak Cyber Partner Program
New service AI-driven application security assessment and monitoring
Related initiative Project Lightwell ($5B investment)
Access model Read-only, bounded-execution security harness
Delivery Managed enterprise service, available now

Mark Hughes, Global Managing Partner for Cybersecurity Services at IBM Consulting, said: "Attackers are already using AI to probe, exploit, and scale threats at machine speed. Defenders need the same advantage, with the security and control enterprises require." OpenAI's Chief Information Security Officer, Dane Stuckey, added that "security is central to realizing the benefits of advanced AI," and that the partnership aims to accelerate defensive security workflows for enterprises, governments, and other organizations as they adopt AI under the trust and compliance their environments demand.

Security teams evaluating AI-driven vulnerability scanning tools should treat access scope as the first design decision — define repository permissions, execution boundaries, and governance policy before rollout. IBM's "read-only, bounded execution" model is a useful baseline for vendor evaluation criteria.
This partnership reflects a broader trend: as attackers increasingly weaponize AI for faster reconnaissance and exploitation, defenders are racing to match that speed with their own AI-powered tooling. Expect more consulting and infrastructure firms to pair up with frontier model providers on similar defensive security offerings.

Why It Matters

The announcement underscores how frontier AI models are moving from productivity tools into core security infrastructure. As AI-assisted attacks accelerate, defenders face mounting pressure to match that pace — and this partnership is one of the more concrete enterprise responses so far, pairing a named, shippable product (the application security service) with a much larger supply-chain security commitment (Project Lightwell). IBM and OpenAI both signaled that further integrations are planned, suggesting the collaboration will expand into additional security domains over time.

  • IBM joins OpenAI's Daybreak Cyber Partner Program and ships a new AI-powered application security service
  • The service prioritizes exploitable vulnerabilities using AI analysis, going beyond traditional code scanning
  • Tied to the $5 billion Project Lightwell initiative for open-source supply chain security
  • Read-only, bounded-execution access preserves security controls while enabling large-scale analysis