Raw capture — raw/ai-productivity-paradox.md
url: https://www.svpg.com/the-ai-productivity-paradox/ title: The AI Productivity Paradox author: Marty Cagan date_fetched: 2026-07-25 date_published: 2026-07-23 site: Silicon Valley Product Group (SVPG)
Marty Cagan addresses a growing industry recognition that AI is accelerating output without corresponding improvements in business outcomes — a phenomenon now called the "AI Productivity Paradox."
From McKinsey Quarterly: "Adoption of generative and agentic AI is growing, investment is accelerating, but sustained impact on performance is elusive."
From Atlassian's State of Teams 2026 Report: "89% of executives say AI has increased the speed of work, but only 6% feel confident they can point to specific organization-wide AI ROI."
From AI product leader Hilary Gridley: "It's never been faster to build, which means it's never been easier to run 10 times faster in the wrong direction."
From Chip Huyen (author of AI Engineering): "AI makes building easier, but the hardest part remains knowing what to build."
Cagan argues this isn't really a paradox to those who've studied the gap between output and outcomes long before AI. Most teams are using AI to "speed up their old, project model way of working." The project model's core flaw was never slowness; it was being "designed to deliver output, rather than outcomes."
Cagan notes he initially believed generative AI would level the playing field between companies with the best engineers and everyone else. With hindsight, he sees that top engineering talent also correlated with strong product culture. "Rather than closing the gap, the strong product companies are increasing the distance between themselves and the majority of the market."
The piece contrasts two modes:
Many leaders remain convinced that faster building of their ideas will yield results. Cagan suggests we'll have to wait for them to confront the evidence: "the real issue is that their ideas so often prove to be not worth building." For those who embrace the discovery-vs-delivery distinction, though, he states there has "never been a better time" to be creating technology-powered products.