OpenAI расширяет программу Daybreak и выпускает ИИ-модель для кибербезопасности

OpenAI on August 10, 2026, announced the expansion of the Daybreak program — the company's AI cyber defense initiative — and introduced a new specialized model for cybersecurity tasks, as reported by TechCrunch.

What Is the Daybreak Program

Daybreak is an OpenAI initiative aimed at applying language models in enterprise cyber defense. Along with the program expansion, the company released a specialized cyber model — the first publicly announced OpenAI product specifically for this vertical. Technical specifications and commercial access terms were not disclosed at the time of the announcement.

Demand for such tools is understandable: security teams (SOC, Security Operations Center) process thousands of alerts daily, classify incidents, write threat detection rules, and analyze malicious code. Language models address precisely these routine analytical tasks — and this area is already being actively developed by OpenAI's competitors.

Why AI Attacks Have Become a Systemic Threat

"Attacks using AI are multiplying," — this is how

TechCrunch frames the context for the Daybreak launch.

The availability of language models has dramatically lowered the entry barrier for malicious actors. AI is used today in cyberattacks across four main scenarios:

• Personalized phishing — emails tailored to publicly available data about the target, indistinguishable from real corporate correspondence
• Automated malicious code — generation of malware and its variations targeting specific vulnerabilities that bypass signature databases
• OSINT reconnaissance — automated collection and structuring of data about a target's corporate infrastructure
• Voice deepfakes — cloning executives' voices for telephone social engineering

The speed of attack preparation using AI is significantly higher than manual methods: what used to take days now takes hours. This changes the equation: SOCs need tools that operate at the same speed.

How

OpenAI Is Entering Competition in the Cybersecurity Market

OpenAI has two main competitors in the AI market for enterprise security. Microsoft Security Copilot, launched in 2023, is integrated into Azure Sentinel and Microsoft Defender and has access to threat data from millions of corporate clients. Google Security AI Workbench relies on Mandiant — one of the world's largest threat intelligence providers.

OpenAI's position is built on flagship GPT-series models, the ChatGPT Enterprise corporate network, and experience fine-tuning on professional data. A specialized cyber model trained on CVE databases, malware samples, and MITRE ATT&CK framework tactics could potentially analyze threats more accurately than a general-purpose universal assistant.

What This Means

OpenAI is moving from a universal AI assistant to a platform of vertical solutions: cybersecurity is joining the ranks of medicine, law, and education. The real effectiveness of the Daybreak model will emerge during the first corporate pilots — that is where it will become clear whether it can compete with mature products that have been integrated into corporate infrastructure for several years.