Project Perception - Cybersecurity at the speed of AI
John Savill gives an overview and demo of Project Perception, focusing on how AI-driven workflows and agent-style collaboration can speed up cybersecurity operations across red/blue/green team activities and playbook-driven response.
Overview
The video introduces Project Perception and frames it as an evolution in how security teams operate when AI is part of the workflow.
Cybersecurity evolution
- Savill outlines how security operations are changing and why faster decision-making and response loops matter.
- The discussion sets up the need for tooling that can coordinate work across different security roles.
MDASH and Perception
- Savill introduces MDASH and then moves into Perception, positioning Perception as the core concept/tool being demonstrated.
- The focus is on how Perception is used in practice rather than purely theoretical security concepts.
Agents and team roles (red, blue, green)
- The video discusses agents and how they map to different security functions.
- It breaks down responsibilities and perspectives across:
- Red (offensive/adversary simulation)
- Blue (defensive operations)
- Green (supporting/adjacent operational role as presented in the video)
- Emphasis is placed on how these roles can work together more effectively.
Using Perception and playbooks
- Savill walks through using Perception and highlights playbooks as a key mechanism for repeatable response and operationalization.
- The demo portion shows Perception “in action” and then expands to the idea of Perception being usable broadly (“everywhere”).
Relationship to Security Copilot
- The video explicitly raises the question of Security Copilot in relation to Perception, as part of positioning where Perception fits in the Microsoft security/AI landscape.