● Covered by 6 sources · 6 reportsMedium impact6 neutral
Microsoft Launches MAI-Cyber-1-Flash and Perception for AI Cybersecurity
Key points
- MAI-Cyber-1-Flash is Microsoft's first cybersecurity-specific AI model.
- Project Perception is an agentic platform for automating security workflows.
- The tools are designed for identifying and remediating software vulnerabilities.
- Microsoft claims 96% on CyberGym benchmark, beating competitors.
- The new configuration reportedly cuts costs by 50% compared to previous setups.
New AI Cybersecurity Offerings
Microsoft has launched MAI-Cyber-1-Flash, its first AI model specifically trained for cybersecurity, alongside Project Perception, a new AI cybersecurity platform. These tools are intended to help organizations identify and reduce exposure to security risks, particularly in software vulnerability analysis.
MAI-Cyber-1-Flash is integrated into MDASH, Microsoft's multi-agent harness for finding and fixing software vulnerabilities. Project Perception is designed to deploy teams of agents to assist with and automate various security workflows, including bug detection and remediation, and can also integrate with MDASH.
Performance and Cost Efficiency
Microsoft states that MAI-Cyber-1-Flash, when combined with GPT-5.4 within the MDASH harness, achieved a 96% score on the CyberGym benchmark. This performance is claimed to surpass other models, including Anthropic's Mythos 5, Google's 3.5 Flash Cyber, and OpenAI's GPT-5.5 Cyber and GPT-5.6 Sol.
The company also claims that this configuration reduces operational costs by approximately 50% compared to Microsoft's previous production setup. MAI-Cyber-1-Flash is designed to handle up to 90% of MDASH tasks, reserving larger, more costly models like GPT-5.4 for the remaining 10% of complex tasks.
Agentic Defense Platform
Project Perception is an agentic security system that coordinates different types of AI agents. These include "red team" agents for identifying potential compromise paths, "blue team" agents for investigating and triaging risks, and "green team" agents for remediation and hardening defenses. Project Perception is entering public preview.
Availability and Context
MAI-Cyber-1-Flash is available to approved MDASH customers through an Azure AI Foundry private preview and is not offered as a standalone public model or general-purpose API. The CyberGym Level 1 test, on which the performance claims are based, measures an AI system's ability to reproduce known vulnerabilities from descriptions and source code, not blind vulnerability discovery or patch correctness.
The launch follows the return of Hayete Gallot to Microsoft as executive vice president of security, marking a significant move for the company's cybersecurity business.
✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →
The daily brief
One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.
One email a day. Unsubscribe in one click, any time.
Today's brief
Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.
▶ Play today's brief
New every morning, and the back catalogue is archived by date.
Primary sources
CVE CVE-2026-505229.8 CRITICALHow outlets covered it
Microsoft has launched MAI-Cyber-1-Flash, its first AI model specifically designed for cybersecurity, which reportedly outperforms competitors in vulnerability discovery. This model is integrated into Microsoft's MDASH system to enhance vulnerability identification and remediation, offering significant cost savings.
Microsoft introduced MAI-Cyber-1-Flash, a new cybersecurity-specific AI model, into its MDASH vulnerability identification and remediation platform. This integration, combined with GPT-5.4, scored 95.95% on the CyberGym Level 1 test and reduced operational costs by 50% compared to previous MDASH configurations. The development provides MDASH customers with a more efficient and cost-effective tool for identifying known vulnerabilities.
Microsoft announced MAI-Cyber-1-Flash, a new AI model designed to identify cybersecurity vulnerabilities, which will be integrated into Project Perception. This model reportedly outperforms competitors on the CyberGym benchmark at 50% of the cost, marking Microsoft's first dedicated generative AI model for cybersecurity and a significant move to rejuvenate its cybersecurity business.
Microsoft launched new AI tools, including MAI-Cyber-1-Flash, to automate the identification and reduction of security risks, specifically for software vulnerability analysis. This development matters as it integrates AI into critical security processes, potentially enhancing threat detection and remediation for Microsoft's extensive customer base.
Microsoft introduced MAI-Cyber-1-Flash, a new compact AI cybersecurity model, and Project Perception, an agentic defense platform. This aims to reduce enterprise security costs by efficiently handling most tasks with a smaller model while escalating complex issues to larger frontier models like OpenAI's GPT-5.4.
Microsoft introduced MAI-Cyber-1-Flash, its first cybersecurity-specialized AI model, designed to identify vulnerabilities in codebases. Concurrently, the company launched Perception, an AI cybersecurity platform that uses agent teams to automate security workflows, including bug detection and remediation. These releases aim to enhance enterprise defense capabilities against AI-powered cyberattacks.