Key Takeaways

  • Economists have a track record of wildly misjudging automation's economic impact, going back to David Ricardo's initial fears about the Industrial Revolution in 1820.
  • Predicting AI's long-term effects is nearly impossible because we lack data on consumer demand elasticities for new AI-driven goods or how jobs will truly evolve.
  • Trying to forecast without this crucial data is like a 1400 Mongolian economist guessing what will be scarce in a modern economy, according to Dwarkesh Patel and Phil Trammell.
  • Instead of making specific predictions, focus on mapping out multiple potential scenarios for your business, each defined by different dimensions of scarcity.
  • Alex Imas proposes a “Manhattan Project for data” to gather the economic insights necessary to understand AI's true structural shifts.

Forecasting is a Fool's Game (Especially with AI)

Founders and builders live in a world obsessed with predictions: market sizes, user growth, technological timelines. But when it comes to the macro impact of automation, especially AI, history shows our crystal balls are consistently cloudy. Economist Alex Imas points to the 1820s as a prime example. “We have been famously terrible at forecasting,” Imas notes, referencing David Ricardo's initial pessimistic views on the Industrial Revolution. Ricardo, a titan of economic thought, worried about widespread unemployment. What he missed was the “economics of structural change,” where automation made goods cheaper and created entirely new industries and jobs.

This isn't to say AI will mirror the Industrial Revolution exactly, or guarantee full employment. As Imas clarifies, “I'm not using this anecdote to say this is what's going to happen now... I'm using it to say it's really hard to make predictions.” The lesson isn't what happened, but that even the smartest minds fail to foresee complex, cascading economic shifts.

The Data Deficit: Why We're Flying Blind

The reason AI's economic future remains a black box isn't a lack of smart people, but a gaping void in hard data. Right now, we simply don't have enough information to make reliable long-term forecasts. “We don't have any data,” Imas states directly. He argues for a "Manhattan Project for data" specifically to measure "consumer demand elasticities." We don't know how elastic demand will be for goods and services produced by advanced AI, or how new forms of work will generate different value. Without this, any prediction is just an educated guess—or worse, a shot in the dark.

Dwarkesh Patel underscores this challenge with a vivid analogy. He points to Phil Trammell's idea of “some Mongolian economist sitting around in 1400 thinking about what will be scarce and the limits of that kind of analysis.” A 15th-century thinker couldn't possibly grasp the scarcities of modern industrial or information economies. Similarly, trying to project AI's impact without current, relevant economic data is a futile exercise.

Map Scenarios, Don't Predict the Future

If precise predictions are a dead end, what's a founder to do? The answer is to ditch the crystal ball and embrace scenario planning. Instead of betting on a single future, prepare for several plausible ones. Imas advises, “What is really useful is to think about the potential scenarios, map them out, and say what dimension of scarcity will generate each scenario.”

This means identifying the core uncertainties: Will computational power remain scarce, or become infinitely abundant? Will human attention be the ultimate bottleneck? Will novel materials be hard to find, or easily synthesized? Each answer helps sketch a different future, allowing you to build resilience and flexibility into your startup's strategy. It's about understanding the boundaries of possible realities, rather than chasing a phantom forecast.

What to Do With This

For your next major strategic decision—whether it's launching a new product, raising capital, or re-architecting your team—stop asking, "What will happen?" Instead, map out 2-3 distinct future scenarios for your industry. For each scenario, identify the single most critical resource (e.g., AGI access, specialized human talent, consumer trust) that becomes either extremely scarce or incredibly abundant, then draft a one-page operating plan for how your business would thrive in that specific reality.