We're about to witness something remarkable in entrepreneurship—a surge of new company formation driven by artificial intelligence. I'm calling it a Klepperian Explosion, after the late economist Steven Klepper, whose work on entrepreneurial spinoffs helps explain what's coming.
The Cambrian Parallel
During the Cambrian period, roughly 540 million years ago, Earth experienced an explosion of new life forms. Over a geologically brief span of perhaps 20 million years, nearly all major animal phyla appeared in the fossil record—an unprecedented burst of biological innovation after billions of years of relatively simple life. Something similar is about to happen in the business world—not with organisms, but with AI-native companies.
Klepper's Insight: Disagreement as a Driver of Entrepreneurship
Steven Klepper, a brilliant academic whose work shaped my thinking during graduate school (studying under Gwendolyn Lee in her courses on strategic management and entrepreneurship), developed a theory that explains why spinoffs happen. His research found that disagreements are a massive driver of new company formation. Here's the logic: Employees with tacit knowledge—the people actually doing technical work, meeting with customers, and tracking industry trends—often understand opportunities better than their managers do. When these employees bring innovations or market opportunities to leadership, they sometimes get turned down. As Klepper documented in his 2007 study on the automobile industry:
"Disagreements arise because incumbent management has a limited ability to recognize employees with superior ideas and/or abilities. When the disagreements are severe enough, employees leave to found spinoffs."
The pattern repeated across industries. His 2010 research identified:
"Disagreements among leading decision makers concerning fundamental ideas about technology and management that prompt dissidents to leave and start their own firms."
The classic example: Intel spun out of Fairchild Semiconductor over disagreements about compensation and control. The "Fairchildren," as they're called, went on to reshape the technology industry.
Why AI Creates the Perfect Conditions for Spinoffs
Artificial intelligence is exceptionally difficult to integrate into established companies. The conditions are ripe for disagreement:
Employees are conflicted. My own research analyzing sentiment in Reddit occupational forums around the release of AI models found that professionals experience what we call "high-intensity ambivalence"—feeling equal parts excited and scared about generative AI simultaneously. They see the potential but worry about replacement.
Management is cautious. Top management teams often don't understand the technology, can't envision implementation, or are (reasonably) concerned about real downsides like hallucinations and reliability.
The evidence of productivity gains is overwhelming. Study after study shows that generative AI makes individuals and teams dramatically more productive.
These mixed signals create the classic Klepperian setup: knowledgeable employees who understand what AI can do, working for companies that won't act on that understanding.
The Explosion
I have a personal rule about not predicting the future when it comes to AI—the technology moves too fast and too unpredictably. But I think you can follow the logic here.
If you combine what we know about how transformative technologies take years to show up in organizational performance with Klepper's strategic disagreement theory of spinoff formation, a pattern emerges. AI enthusiasts who deeply understand customers, who know how to productize innovations, and who see what's possible with these tools will bring their ideas to leadership. Many will be turned down by management teams that don't fully grasp the technology or are concerned about its downsides.
And when that happens? Klepper's theory suggests those high performers will leave to form AI-native companies instead.
Whether this becomes a trickle or a flood remains to be seen. But the conditions for a Klepperian Explosion are clearly in place.