Microsoft Launches Scout, AI Personal Assistant Built to Make Users 'Addicted'
Key Takeaways
- ▸Microsoft announced Scout, an AI personal assistant integrated with Microsoft 365 built on OpenClaw, targeting knowledge workers in finance, legal, operations, and HR
- ▸Leaked internal strategy documents reveal an explicit three-phase plan where Phase 1 is explicitly titled 'Make people addicted,' with the goal of building dependency through daily usage
- ▸Scout is an evolution of internally-piloted tool ClawPilot, which has been tested with Microsoft employees since March as part of Project Lobster
Summary
Microsoft officially announced Scout on Tuesday, a personal AI assistant integrated into Microsoft 365 that is built on OpenClaw and designed to help knowledge workers manage calendars, triage inboxes, file expenses, and run recurring workflows. The product is the result of Project Lobster, an internal initiative led by executives Omar Shahine and Jakob Werner, which has been piloting the tool as "ClawPilot" since March.
What distinguishes this product launch is a leaked internal strategy document titled "ClawPilot: Overview and Plan with Project Lobster," obtained by 404 Media, which explicitly outlines a three-phase rollout plan with the first phase explicitly titled "Make people addicted." The document states the goal is to "build the skill and tool ecosystem that makes people depend on it daily" through sustained shipping and user base growth. The pilot program with Microsoft employees has already shown "daily usage with high retention and intensity of usage," according to internal metrics.
The revelation has sparked concern among some Microsoft employees, with one calling the addiction language "very troubling" and questioning why any product should explicitly make addiction a core component of its build strategy. Others have noted the paradox that major tech companies may share similar unstated goals but rarely make them explicit in official documentation.
- The leaked addiction-focused strategy has raised ethical concerns among Microsoft employees about intentionally designing products to create user dependency
Editorial Opinion
The explicit goal to engineer addiction represents a troubling escalation in product design ethics. While driving engagement has long been standard practice in consumer technology, Microsoft's willingness to codify addiction as a strategic objective—and put it in writing—exposes the darker assumptions behind modern AI products. This leaked document should prompt regulatory scrutiny: if Microsoft is openly building addictive AI agents, what are other AI companies doing quietly behind closed doors?
