The McKinsey Podcast.
This McKinsey podcast article, featuring Alexis Krivkovich, Lucia Rahilly, and Roberta Fusaro, examines why organizations struggle to capture value from artificial intelligence. While companies are rapidly adopting AI, most are not seeing measurable returns. In fact, more than 80% report little impact on their bottom line. The authors argue that the main challenge is not the technology itself. Instead, the difficulty lies in redesigning workflows, leadership, and organizational structures for an “agentic” future.
The AI paradox and unrealized value
The article introduces a central paradox. Companies expect transformative gains from AI and continue investing heavily. However, results remain limited because adoption is often fragmented. Many firms focus on isolated use cases rather than systemic transformation.
Agentic AI raises the stakes further. It enables autonomous systems that can perform complex, multi-step tasks. As a result, the potential value is significantly higher. Yet, realizing this value requires rethinking entire business models. For example, near-zero marginal costs and highly personalized services could reshape competition across industries.
Rethinking workflows and organizational design
A key argument is that value emerges when organizations redesign workflows end to end. Point solutions improve individual tasks, but they rarely generate large-scale impact. By contrast, reimagined workflows integrate multiple functions and allow AI agents to coordinate activities across processes.
This shift also affects team structures and decision-making. Organizations may reduce layers of hierarchy and enable faster decisions. At the same time, employees will need to collaborate with AI systems in new ways. The concept of “humans above the loop” illustrates this change. Instead of executing tasks, workers increasingly supervise, evaluate, and guide AI-driven processes.
Skills, leadership, and cultural transformation
The transition to an agentic organization requires substantial changes in skills. Around 75% of roles are expected to be reshaped in the near term. Consequently, organizations must invest heavily in capability building. Technical expertise remains important, but demand for judgment, problem-solving, and strategic thinking is increasing.
Leadership also plays a critical role. Executives must rethink how they allocate their time and lead transformation. Moreover, they need to balance experimentation with risk management, especially given concerns about AI reliability. Trust remains a major barrier, as early use cases have revealed limitations such as hallucinations.
Cultural factors are equally important. Organizations that promote curiosity, continuous learning, and cross-functional collaboration are better positioned to adapt. Ultimately, the article emphasizes that change management is no longer episodic. It has become a continuous process in an environment of constant technological evolution.
Reference
Krivkovich, A., Rahilly, L., & Fusaro, R. (2026, April). AI is everywhere. The agentic organization isn’t—yet. McKinsey & Company. https://www.mckinsey.com/capabilities/