Sophos reports 96% faster threat investigations with OpenAI
Sophos says agents built through OpenAI's Daybreak program have cut threat investigation time by 96% and now resolve about half of its managed detection and response cases without human handling.
Security vendor Sophos says it has sharply cut threat investigation times using OpenAI models accessed through Daybreak, OpenAI's cybersecurity program. The figures come from a customer story published by OpenAI.
The company reports three headline figures:
- 96% reduction in investigation time using OpenAI models
- 89 seconds average response time for cases handled by AI agents
- 52% of managed detection and response (MDR) cases resolved end-to-end by AI
A Sophos representative, who is not named in the published material, described the arrangement as a division of labour. Sophos supplies "the domain expertise and the know-how to combat attacks at scale," and Daybreak provides "frontier intelligence" that lets the company extend that expertise to all of its customers.
"Because of the agents we've been able to build through the Daybreak programme, the average response time for cases using those agents has fallen to about 89 seconds," the representative said. "About half of the cases we handle are now being automated by agents we developed using the Daybreak models."
The published material does not say which OpenAI models Sophos uses, how the 96% baseline was measured, which kinds of cases the agents handle on their own, or how automated resolutions are checked for accuracy. It also gives little detail about how Daybreak itself is structured.
All of the figures are self-reported by the two companies and have not been independently verified. Even so, they are a rare quantified example of agentic AI running in production security operations, where speed of response directly limits how much damage an attacker can do.