How Leaders Can Use AI to Solve Real Business Problems

Josh Tyrangiel, a staff writer at The Atlantic and author of AI for Good, argues that most organizations are approaching artificial intelligence backward. They are "solving for AI"—pushing adoption for its own sake—rather than using AI as a tool to solve a specific, well-understood business problem. Drawing on years of reporting, Tyrangiel offers a sober, practical framework for leaders: start with the problem, not the technology; embed domain expertise into every pilot; and treat AI as a management challenge, not just a technical one.

The Hype Cycle and the "Drunkenness" Problem

Tyrangiel describes the current AI landscape as a "classic spin cycle" where everyone is "a little bit drunk." The technology is genuinely impressive, but the hype—fueled by marketing, venture capital pressure, and the labs' need to prove profitability—creates a dangerous illusion. The labs, which are for-profit companies calling themselves "labs" to borrow scientific credibility, are deeply in debt and need to convince leaders that AI is indispensable. Tyrangiel warns that the people who build these models are often terrible at understanding human systems, customers, or the "soft skills" critical to business success. The result is a push to "get on the boat" that the labs themselves financed, not a neutral assessment of what actually works. Leaders must resist this pressure and instead ask a