Samsung SDS said on Sept. 8 it will join OpenAI’s Daybreak Partner Program as a pilot partner, a programme run with various organisations to strengthen security capabilities using AI. It said it will expand its AI-based cybersecurity business in earnest.

Daybreak uses OpenAI frontier models and focuses on improving the entire security process, including vulnerability discovery, verification of real-world impact, risk prioritisation, generation and testing of security patches, and confirmation of final fixes.

Samsung SDS plans to participate as a Daybreak pilot partner and provide AI-based cyber defence services to corporate customers by combining global AI security technology with its expertise in building and operating large-scale IT systems and its industry-specific specialisation.

Samsung SDS plans to closely support the full security process through field-focused FDEs and security experts, from identifying security tasks at customer sites to linking and optimising security solutions and security monitoring platforms.

A Samsung SDS official said the company would actively support efforts to improve corporate customers’ security levels by combining OpenAI frontier AI models with Samsung SDS’s IT service capabilities as a Daybreak pilot partner. The official said Daybreak would go beyond simply finding vulnerabilities by verifying the likelihood of actual attacks and business impact so customers can respond quickly, starting with the most important risks.

Earlier, OpenAI expanded and reorganised Daybreak into two parts, Daybreak Blue and Daybreak Red, and decided to provide a new model, GPT-5.6-Cyber, to Red users.

According to OpenAI, Daybreak Blue supports access to OpenAI’s advanced general-purpose models with safeguards adjusted for defensive security work. OpenAI recommends that most organisations start with Daybreak Blue.

Daybreak Red provides access to cybersecurity models trained for security testing, vulnerability research and exploit verification. Users at this stage can also use GPT-5.6-Cyber.