McKinsey’s New AI Report: Productivity Payoff Real But Conditional
May 1, 2026 – 10:25 am
The firm’s new report, "AI productivity gains and the performance paradox", concludes that most current AI applications "accelerate existing work" without significantly altering workflows. McKinsey aims to achieve 1:1 parity between its 40,000 human consultants and 40,000 AI agents by year-end.
The AI Paradox
According to the report, the corporate world is facing an "AI paradox": increasing adoption and investment in generative and agentic AI has not led to substantial improvements in performance. The authors compare this to the introduction of electricity in factories: initially, businesses replaced steam engines with electric motors, achieving efficiency gains but preserving the existing layout. It was only later, when motors allowed for the rearrangement of machines around workflows, that truly transformative changes occurred.
Key Takeaways from McKinsey’s Report
-
AI as a Tool: Most current AI applications are "tools that accelerate existing work" but "largely preserve underlying workflows."
-
Redesign for AI: Larger productivity gains will only come when organizations redesign their processes around AI, not simply add it on top.
-
Historical Analogy: The introduction of electricity in factories serves as a central historical analogy to illustrate the need for fundamental change to unlock AI’s full potential.
-
Three Recommendations:
- Assess how AI will reshape industry profit pools
- Build or strengthen AI-powered competitive moats
- Turn speed into a structural advantage
Examples
The report cites examples of AI applications that fall into two categories:
- Work Acceleration: JPMorgan Chase’s real-time AI fraud detection, BMW’s computer vision quality inspection, and Siemens’ AI-coordinated predictive maintenance.
- Deep Process Redesign: (Examples not explicitly stated in the provided text)
The Growing Disparity
The report comes at a time when the gap between AI investment and demonstrable returns has become evident. The Federal Reserve Bank of St. Louis found a 1.9% excess cumulative productivity growth since ChatGPT launched in November 2022, while JPMorgan warned that substantial continuous revenue would be needed to justify current AI capex. MIT Media Lab research indicates that 95% of organizations see no measurable returns from AI adoption, and Deloitte’s 2026 report highlights similar concerns.