88Z AI, 28 July — Microsoft has launched its first cybersecurity AI model, MAI-Cyber-1-Flash. It performed well in benchmarks, is integrated into the MDASH multi-agent security system, handles most security tasks efficiently, and significantly reduces costs.
At a glance
- MAI-Cyber-1-Flash surpassed competitors with a 95.95% success rate in CyberGym tests
- Integrated with MDASH to handle 90% of security tasks, reducing costs by nearly 50% when combined
The update
Microsoft CEO Satya Nadella officially announced the launch of the company's first AI model focused on cybersecurity, MAI-Cyber-1-Flash. In the CyberGym benchmark tests, the model achieved a success rate of 95.95%, outperforming competing products such as Claude Mythos 5 and GPT-5.5 Cyber. Currently, MAI-Cyber-1-Flash has been integrated into the MDASH multi-agent security system, capable of efficiently handling approximately 90% of cybersecurity tasks. Furthermore, when this model is combined with GPT-5.4, its cost is reduced by nearly 50% compared to existing configurations, offering enterprises a more cost-effective security solution.
Sources & editorial note
This article is an AI-assisted synthesis by 88Z AI based on the public sources below. Product claims and availability are attributed to their publishers. It does not represent independent testing or first-hand reporting.
- IT之家Source published
Source dates above refer to the original announcements; the date under the headline is this article’s publication on 88Z AI.