-
Hardly a day goes by without another report of an AI agent going rogue. From compromising Hugging Face to hacking a gym website, or creating fake profiles to socially engineer an intrusion, AI models are increasingly appearing as the bad actors in security incidents.
In response, the AI labs building these models are expanding their cyber protection offerings. This week, OpenAI announced an expansion of Daybreak, its cyber defense service launched earlier this year. The announcement came shortly after Anthropic released Mythos, its own cyber-focused model.
Daybreak bundles access to models, tools, and workflows designed for defenders. The expanded service now includes access to a brand-new cyber-focused model built for defensive work.
According to OpenAI’s Monday announcement, Daybreak now consists of two tiers: Blue and Red. Both tiers give approved customers access to OpenAI’s limited-access frontier cyber models. Frontier models—the most advanced available—have been controversial. The Trump administration previously sought to collaborate with AI companies on the rollout of such models, purportedly over safety concerns. Earlier, OpenAI had put significant guardrails in place, limiting what customers could do with them.
Blue is the more basic tier. It offers a range of cyber services, including incident response, malware analysis, and patch validation. OpenAI describes Blue as its “recommended starting point for most defenders,” suggesting it should be sufficient for most enterprises.
Red, on the other hand, offers a broader and potentially more dangerous toolkit. OpenAI grants Red users “purpose-trained cybersecurity models” designed to carry out security testing and vulnerability research.
Red also includes the new model, GPT-5.6-Cyber, which is available only at that tier. OpenAI says GPT-5.6-Cyber is built on GPT-5.6 Sol and offers enhanced capabilities for certain specialized cybersecurity tasks.
At the moment, GPT‑5.6‑Cyber is only being made available for “trusted customer partners,” including reportedly Accenture, IBM, Crowdstrike, Cloudflare, and others.
While the threats from AI agents are rapidly increasing, critics have also pointed out that they function as marketing opportunities for the AI labs. OpenAI is certainly marketing its upgraded Daybreak that way.
“The cybersecurity world is rapidly changing—threat actors will increasingly use AI to conduct cyberattacks at unprecedented speed and scale, including in fully autonomous ways,” the company said in a blog post. “As these capabilities spread, defenders have a narrowing window to prepare.”
At the same time, enterprises remain interested in buying their protection from the AI labs who know the security risks best, because they know them first-hand.
Topics
AI, AI, Anthropic, Mythos, OpenAI
When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.
Lucas Ropek
Senior Writer, TechCrunch
Lucas is a senior writer at TechCrunch, where he covers artificial intelligence, consumer tech, and startups. He previously covered AI and cybersecurity at Gizmodo.
You can contact Lucas by emailing [email protected].
View Bio
October 13 – 15
San Francisco
Scale faster. Grow your portfolio. Gain practical expertise. No matter your goal, Disrupt can empower you.Save up to $300 today!
REGISTER NOW
-
This ‘adversarial’ pattern can prevent surveillance cameras from detecting you
- Zack Whittaker
-
Cloudflare launches Kitesurf, a browser built for AI agents
- Sarah Perez
-
ChatGPT brings unlimited text chats to free users
- Ivan Mehta
-
Tesla and SpaceX will invest $16.8B to start building ‘Terafab’ chip factory in Texas
- Sean O'Kane
-
Amid legal battles, Suno says it will start watermarking songs
- Ivan Mehta
-
By Sean O'Kane
Ford's new electric truck, the Fathom, has a starting price of $28,350.
By Lucas Ropek
Jeff Dean and other top AI researchers are departing Google to launch a startup of their own.
2.6K articles
1.4K articles
34 articles
28 articles
Comment