Your security vendor gets the frontier cyber model, you get the findings

Selected red team specialists can now use OpenAI’s cyber models to find and exploit weaknesses in client applications and infrastructure. Those clients never get the models themselves. That split is the design of the Daybreak Cyber Partner Program, which OpenAI expanded on August 10: access to the underlying models stays with the approved partner and is not transferred directly to the customer.

Approved partners choose between Daybreak Blue and Daybreak Red, reached through Daybreak Access, depending on their needs and the work involved. Blue supports a broad range of defensive security workflows. Red covers more specialized and closely governed work, including red teaming and penetration testing, where a hired team breaks into a customer’s systems to show what a real intruder could reach.

Sixteen companies are named. Nine are security and services firms: Accenture, IBM, Capgemini, Cognizant, EY, KPMG, PwC, NCC Group and SpecterOps. Seven are technology partners: Palo Alto Networks’ Unit 42, CrowdStrike, Cisco, Sophos, Akamai, Fortinet and Cloudflare.

What partners can do with the models

Depending on the engagement, partners can help with vulnerability discovery and validation, red teaming, penetration testing, incident response, and remediation across complex enterprise systems. Finding a flaw and writing it up protects nobody. The work that counts comes after: whether the weakness can actually be exploited, which systems are at risk, what the fix is, and whether it reaches production. Partners already know their customers’ systems and how their security teams operate, which is the argument for routing the models through them rather than shipping them to every company that wants one.

Milan Patel, global head of MDR services at Sophos, runs the outsourced monitoring and response side of that business. He describes the arrangement as delivering “frontier-grade defense at scale to organizations unable to deploy these models on their own.”

Safeguards vary by engagement

Safeguards can include identity verification, defined testing scopes, logging, monitoring, and human oversight, depending on the work. No fixed set applies everywhere. Partners and their customers define the boundaries of each engagement, review findings, and apply the partner’s judgment before action is taken. The controls exist, and which ones you get is a matter for the contract.

Download: The high-performance team playbook