OpenAI expands cyber defense service as AI-driven attacks rise

As reports of AI agents acting maliciously become more frequent - such as compromising platforms, breaching websites, or creating fake profiles to enable intrusions - the labs behind these models are stepping up their defensive offerings. This week, OpenAI announced an expansion of Daybreak, its cyber defense service launched earlier this year, shortly after Anthropic introduced its own cyber-focused model, Mythos.

Daybreak bundles access to models, tools, and workflows for defenders. The update introduces two tiers: Blue and Red. Both give approved customers access to OpenAI's frontier cyber models, which have drawn controversy. The Trump administration previously sought collaboration with AI companies on rolling out such models, citing safety concerns. OpenAI had earlier placed strict limits on how customers could use them.

Blue is positioned as the recommended starting point for most defenders, offering services like incident response, malware analysis, and patch validation. Red provides a broader toolkit, including purpose-trained cybersecurity models for security testing and vulnerability research. Red also includes the new model, GPT-5.6-Cyber, built on GPT-5.6 Sol, with enhanced capabilities for specialized tasks. It is currently limited to trusted partners, reportedly including Accenture, IBM, Crowdstrike, and Cloudflare.

Critics note that rising AI threats also serve as marketing opportunities for labs. OpenAI is promoting Daybreak accordingly, stating that threat actors will increasingly use AI for attacks at unprecedented speed and scale, leaving defenders a narrow window to prepare. Enterprises, however, remain interested in buying protection from the labs that understand these risks firsthand.

Source: TechCrunch