Agentic AI: What Leaders Wish They Knew Sooner

2026-06-11 · Source: MIT Sloan Management Review · Field: Business & Management — Corporate Strategy & Leadership, Operations & Process Management, Artificial Intelligence & Machine Learning · Depth: Intermediate, medium

Summary

At the MIT Sloan CIO Symposium, IT and business leaders shared insights on human-agentic AI collaboration, revealing both concerns and opportunities. Thomas Davenport expressed worry about humans becoming cursory auditors due to agents' speed, questioning the viability of "human-in-the-loop" models. Melissa Swift debunked the myth of "magical" agents, emphasizing the need for continuous human checking and re-prompting. George Westerman noted that many "agents" are not sophisticated, leading to hype, and advocated for outcome-driven process redesign. Monica Caldas detailed deploying "micro agents" in IT operations, stressing deliberate workflow reimagination, clear OKRs, and building trust for autonomous yet human-judgement-integrated systems. Other leaders highlighted the importance of treating agents like employees, training humans to collaborate with AI for higher-level tasks, and viewing human-agent interaction as collaboration to unlock new possibilities and parallel processes, rather than replacement. The consensus leans towards a hybrid model where humans design and decide, while AI implements, with a focus on building trust and effective human-AI handoffs.

Key takeaway

For Directors of AI/ML evaluating agentic AI deployments, recognize that successful integration demands more than simple automation. You must deliberately redesign workflows around desired outcomes, not just existing steps, and establish clear OKRs for agent performance. Prioritize building robust trust frameworks and governance for autonomous agents, ensuring human judgment is incorporated strategically "in the right places" rather than at every step. Focus on training your teams to collaborate with agents, elevating human roles while agents handle repetitive tasks, fostering a hybrid model where humans design and decide.

Key insights

Agentic AI requires deliberate human-AI collaboration, process redesign, and trust-building, moving beyond simple automation.

Principles

• Agentic AI is not "magical"; it requires human oversight.
• Focus on outcome-driven process redesign, not just automating steps.
• Humans should be "in the right places" for judgment, not every step.

Method

Deploy "micro agents" by iterating workflow evolution, reimagining processes with clear OKRs, and building trust and governance for autonomous yet human-integrated systems.

In practice

• Implement clear entry and exit criteria for agent deployments.
• Train human employees to collaborate with AI for higher-level tasks.
• Design processes where humans decide and AI implements.

Topics

• Agentic AI
• Human-AI Collaboration
• Workflow Automation
• AI Governance
• Process Redesign
• IT Operations

Best for: CTO, Executive, AI Product Manager, Director of AI/ML, VP of Engineering/Data, Consultant

Related on AIssential

• From assistants to agents — Agentic AI autonomously executes goals, shifting human-AI collaboration from reactive assistance to proactive, independent action.
• Most Enterprise Agentic Projects Are Doomed, Here's Why — Jess Grogan-Avignon & Jack Wang, Accenture — Enterprise AI success requires adapting organizational speed, finance, delivery, and trust-building to AI's emergent, machine-speed nature.
• Why Good AI Agents Look Less Like Chatbots and More Like Workshops — Reliable AI agents prioritize constraint and human governance over unbridled autonomy to counter "context rot."
• Building AI agents for the enterprise — Embedding agentic AI into enterprise workflows and institutional knowledge drives significant, compounding returns.
• Why Agents Still Need Humans — AI agents increase demand for human expertise by commoditizing routine tasks and creating an infinite backlog of new work.
• Agentic AI: What Leaders Wish They Knew Sooner — Effective agentic AI integration requires strategic process redesign, gradual trust-building, and human adaptation to new collaborative roles.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by MIT Sloan Management Review.