HBR Says Treat AI Agents Like Team Members. We've Been Doing It Since Day One.
Harvard Business Review just published “To Scale AI Agents Successfully, Think of Them Like Team Members”, arguing that AI agents need clear roles, permissions, scope, accountability, and governance. That they should be managed like employees, not software.
We read it and thought: that’s been our operating model since day one.
The HBR Framework
The article, by Rahul Telang and Noman Hydari from Carnegie Mellon and the University of Pittsburgh, makes the case that most organizations fail at AI deployment because they treat agents as tools instead of team members. Their framework includes:
- Clear roles and responsibilities for each agent
- Defined authority levels with escalation paths
- Accountability structures so you know who did what
- Governance policies to prevent agents from going rogue
Sound familiar?
How RelayLaunch Already Does This
At RelayLaunch, every engagement runs on the platform: 16 Rooms with 51 focused specialists. But the key isn’t the number of specialists. It’s how they’re organized.
Clear Chain of Command
Every agent knows its authority level:
- L0: Owner (that’s you, the business owner). Final say on everything
- L0.5: I.S.A., your AI Chief of Staff. Routes tasks, synthesizes decisions
- L1: Constitutional agents. Enforce security, brand, and quality autonomously
- L2: Directors. 16 Room directors, each with a distinct business perspective
- L3: Specialists. 51 domain experts who do the actual work
No agent acts above its authority. No agent pushes to production without approval. Silence isn’t consent.
Intentional Rivalries
Here’s something HBR doesn’t mention but we think is critical: productive tension between agents.
Our department directors have built-in rivalries. Sales wants to close deals fast. Operations demands strategy reviews first. Growth wants to ship now. Quality insists on quality gates.
These aren’t bugs. They’re features. When agents disagree, blind spots get surfaced. When they agree, you can be confident in the decision.
Conviction Scoring
Every recommendation from I.S.A. includes a confidence score. High confidence? She acts. Low confidence? She tells you exactly what information is missing and what would change the answer.
No false certainty. No hand-waving. Just transparent reasoning.
What This Means for Your Business
If HBR is right (and we think they are) then the businesses that win with AI won’t be the ones with the most agents. They’ll be the ones with the best-organized teams.
You don’t need 300 agents. You need 10 well-structured Rooms with clear authority, accountability, and the productive tension to make better decisions than any single agent could make alone.
See Where You Stand
Take our free 60-second business scorecard to see how your current operations measure up. No call required. Instant results.
Or if you’re ready for the full picture, the Concierge Analysis ($1,500) gives you a complete operations diagnostic with a prioritized roadmap and a 90-minute walkthrough.
Either way, the first step is knowing where you are.