--- title: "Rethinking security for the age of AI" url: https://daily.dev/posts/rethinking-security-for-the-age-of-ai-98trns1ah source_url: https://blogs.microsoft.com/blog/2026/07/27/rethinking-security-for-the-age-of-ai type: article source: "Microsoft" published: 2026-07-27T16:35:01.355Z updated: 2026-07-28T22:40:45.511Z tags: ["security", "microsoft", "ai-agents"] reading_time: 7 upvotes: 1 comments: 0 language: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Rethinking security for the age of AI **[Microsoft](https://daily.dev/sources/microsoft)** · 7 min read · 1 upvotes · 0 comments ## Summary Microsoft is introducing Project Perception, an agentic security system designed for the AI era. It coordinates three classes of specialized agents — red team (attack path discovery), blue team (risk evaluation), and green team (corrective action) — forming a closed-loop defense system. Built on a new 'Cyber Stack' architecture, it combines signals, security context, a multi-model AI approach, and actuators to enable continuous, machine-speed threat response. A specialized model, MAI-Cyber-1-Flash, is integrated for software vulnerability management, achieving 96% on the CyberGym benchmark with ~50% cost savings over the current configuration. Project Perception enters public preview on August 3. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: --- Tags: [#security](https://daily.dev/tags/security), [#microsoft](https://daily.dev/tags/microsoft), [#ai-agents](https://daily.dev/tags/ai-agents) [View this post on daily.dev](https://daily.dev/posts/rethinking-security-for-the-age-of-ai-98trns1ah)