OpenAI Expands Daybreak Cybersecurity Program Amid AI Attack Concerns

OpenAI expands its Daybreak cybersecurity initiative, introducing a new cyber-trained AI model.

OpenAI has expanded its Daybreak cybersecurity defense program, rolling out a new cyber-trained AI model to address increasing AI-led attacks. This expansion introduces two tiers, Blue and Red, offering approved customers access to OpenAI’s limited-access frontier cyber models. The Blue tier provides services such as incident response, malware analysis, and patch validation, while the Red tier offers a broader toolkit for security testing and vulnerability research.

The Red tier includes access to the new GPT-5.6-Cyber model, which is built off of GPT-5.6 Sol and designed for specialized cybersecurity tasks. This model is currently available only to "trusted customer partners," including Accenture, IBM, Crowdstrike, and Cloudflare. The expansion comes as intelligence CIOs warn that AI agents could accelerate cyberattacks, intensifying the AI security debate.

OpenAI states that the cybersecurity world is rapidly changing, with threat actors increasingly using AI for cyberattacks at unprecedented speed and scale. The company emphasizes that defenders have a narrowing window to prepare as these capabilities spread. This move aims to provide enterprises with advanced tools to counter evolving AI agent threats.

Find the right AI tool for your business

Chat with Insta and get matched to the right tool in seconds.