Rethinking organizational design in the age of agentic AI | MIT Technology Review
3 min read
Fundamentally, many organizations want to use advanced agentic AI. However, they often just add it to their old ways of working. Consequently, this “sticky tape” approach limits the AI’s true power.
Moreover, leaders need a new plan for organizational redesign. Critically, this means changing their entire system. For example, they must rethink their tech, their teams, and their goals.
Significantly, this shift to agentic business transformation helps AI work with people. Therefore, organizations can move faster. Consequently, they can measure success by real results, not just activity.
| Dimension | Technology Stack | Workforce | Success Metrics |
|---|---|---|---|
| Traditional Model | Linear, application-centric workflows designed for human operation | Hierarchical structure with standardized, siloed tasks and clear delineation between SBUs | Activity/output-focused (e.g., calls handled, reports filed, cost per query) |
| ABT Transformation | AI agents serve as connective tissue across systems, operating at machine speed with multi-system contextualization | Hybrid human-AI teams with redesigned roles, fluid decision-making, and diffused operational accountability | Outcome-focused metrics tied to business value (e.g., contracts reviewed without human escalation, customer retention) |
| Key Challenge | Existing stacks must be fully rearchitected, not merely layered with AI tools (“sticky tape” approach) | Managing trust, explainability, psychological safety, and status dynamics in human-AI teams | Activity metrics become “meaningless or misleading” when AI employees handle 100x the volume of humans |
| Strategic Outcome | Time from business requirement to production workflow drops from months to days | 75% of current jobs will require redesign, upskilling, or redeployment by 2030 (McKinsey) | Enterprise case study showed ROI from agentic AI tripled within two quarters after metric overhaul |
Agentic AI Drives Organizational Change
Systemic Change Required for AI
This indicates a disconnect between organizational ambition and current readiness for agentic AI. Therefore, simply layering AI onto old systems is insufficient. Similarly, many report their operations cannot support the change. Moreover, success requires systems-level redesign. Consequently, enterprises must rethink their technology, workforce, and metrics to fully benefit.
“ABT is something categorically different: It’s the integration of AI agents into the fabric of the organization.”
Ultimately, integrating agentic AI demands a full systems-level redesign. In conclusion, the ABT framework provides a clear path forward. Looking ahead, companies must rewire their tech, workforce, and success metrics. As a result, they can achieve meaningful and lasting benefits. Therefore, it is vital to move beyond superficial AI adoption. Thus, organizations can unlock true competitive differentiation. Hence, embracing this transformation is no longer optional. In summary, the pillars of ABT address core operational shifts. To conclude, leaders should start this essential dialogue now. Finally, proactive change will define future market success.
