Most autonomous vehicle startups make the same critical mistake: they focus on easy solutions rather than the rare, complex driving situations that occur once every million miles. That's the warning from Waymo co-CEO Dmitri Dolgov, who explained in a recent podcast that tech breakthroughs spark hype cycles where startups rush in but rarely last beyond the initial excitement.

What's the Real Difference Between Hype and Sustainable Progress?

Dolgov described a pattern he's observed across technology cycles. When major breakthroughs happen, like the recent advances in large language models (LLMs), a wave of startups enters the market. But most don't survive because they misunderstand what it takes to build a truly reliable product. "It is very easy to get started but very difficult to take it all the way to a real product, full autonomy, and superhuman performance," Dolgov explained.

Dolgov

The key distinction lies in what Dolgov calls the "long tail." These are edge cases, rare driving situations that occur once every million miles that a self-driving system must safely navigate. Big breakthroughs like new AI models change the "early part of the curve," he noted, but not the long tail where real reliability is proven.

"Today, worldwide, somebody loses their life in a crash on our roads every 26 seconds. It's the combination of knowing that the mission is really important and understanding what you're up against, not looking for easy wins or quick solutions or silver bullets, that helps the team have the stamina to go the distance," said Dmitri Dolgov.

Dmitri Dolgov, Co-CEO at Waymo

This perspective reflects Dolgov's deep experience in the field. Before joining Waymo in 2009, he worked on self-driving car technology at Toyota and Stanford University. He shares the CEO title with Tekedra Mawakana.

How Do Successful Autonomous Vehicle Companies Build Lasting Products?

• Focus on Edge Cases: Rather than celebrating early wins on simple driving scenarios, invest heavily in identifying and solving rare, complex situations that occur once every million miles of driving.
• Commit to Long-Term Development: Understand that moving from proof-of-concept to a fully autonomous, superhuman-performing system requires years of sustained effort, not quick pivots chasing hype cycles.
• Align Mission with Reality: Keep the stakes clear: road safety is literally a matter of life and death, with someone dying in a crash every 26 seconds globally, which should drive decision-making away from shortcuts.
• Build Institutional Stamina: Create a culture and team structure that can sustain focus through multiple technology cycles, resisting the pressure to chase trendy breakthroughs that don't address core reliability problems.

Waymo's own trajectory demonstrates this philosophy in practice. The company was founded by Google in 2009 as a self-driving technology company. It launched its first autonomous vehicles in Phoenix in 2020, more than a decade after its founding, and is now available in several U.S. cities, including the San Francisco Bay Area, Los Angeles, Miami, and Nashville. The company's robotaxis are all electric and use artificial intelligence (AI), mapping technology, and sensors to drive autonomously.

Today, Waymo is one of the few autonomous vehicle companies offering robotaxi services in the United States, alongside Tesla and Uber. This exclusivity reflects the difficulty of the problem Dolgov described. While many startups have entered the autonomous vehicle space over the past 15 years, few have achieved the operational scale and reliability needed to deploy commercial robotaxi services.

The lesson extends beyond autonomous vehicles. Dolgov's observation about hype cycles and the "long tail" applies to any emerging technology where early breakthroughs attract capital and talent. Startups that mistake initial progress for a complete solution often burn through funding before solving the hard problems that separate products from prototypes. Those that survive, like Waymo, are typically those willing to invest years in unglamorous work: testing rare scenarios, refining safety systems, and building the operational infrastructure needed for real-world deployment.

For investors, entrepreneurs, and technologists watching the autonomous vehicle space, Dolgov's message is clear: be skeptical of startups that celebrate early milestones as proof of viability. The real test comes later, when the novelty wears off and the work of solving the long tail begins.