The new tiered service gives defenders access to Blue and Red tiers, with Red featuring the new model for trusted partners.
OpenAI expanded its Daybreak cyber defense service into two tiers named Blue and Red. The expansion introduces GPT-5.6-Cyber, a new frontier model for defensive security work. Daybreak bundles access to models, tools, and workflows for defenders.
OpenAI launched Daybreak earlier this year to counter AI agents that compromise services. The Red tier adds GPT-5.6-Cyber for trusted partners including Accenture, IBM, Crowdstrike, and Cloudflare. Anthropic released its rival cyber model Mythos recently.
Blue tier provides incident response, malware analysis, and patch validation as a recommended starting point. Red tier grants purpose-trained models for security testing and vulnerability research. Enterprises must weigh OpenAI’s access limits on frontier cyber models.
Critics say the new tiers double as marketing for AI labs that also build attacking agents. Watch whether OpenAI widens access to GPT-5.6-Cyber beyond current trusted partners. Watch whether Anthropic updates Mythos to match Daybreak’s tiered approach.
What matters
- OpenAI launched Daybreak’s Blue and Red tiers, with Red adding the GPT-5.6-Cyber model for trusted partners.
- Enterprises can now choose a baseline cyber defense tier or a deeper Red tier for advanced security testing.
- Watch how OpenAI gates access to GPT-5.6-Cyber and whether Anthropic’s Mythos responds with similar tiers.
Why it matters
Watch how OpenAI gates access to GPT-5.6-Cyber and whether Anthropic’s Mythos responds with similar tiers.
This GenAI News article was prepared in original wording using reporting and materials published by TechCrunch AI. Source reference: https://techcrunch.com/2026/08/10/as-ai-led-attacks-multiply-openai-launches-a-new-cyber-model/.
Drafted by the GenAI News review pipeline.