Robots are waiting for a ChatGPT moment: Nvidia’s Les Karpas explains why at TechCrunch Disrupt 2026
TechCrunch TechCrunch Events
Nvidia’s Les Karpas is set to explain why robots still haven’t had a ChatGPT moment. The snag: physical AI lacks the giant internet dataset that powered language models.
Based on reporting by TechCrunch, TechCrunch Events — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Robotics has been promised its big breakout for years. Nvidia thinks the missing piece is still the same one: scale. At TechCrunch Disrupt 2026, Nvidia Inception’s Global Head of Physical AI, Les Karpas, will make that case on the Real World AI Stage in a session called Robots Are Waiting for Their ChatGPT Moment. Here Is What Is Standing in the Way.
The comparison to ChatGPT is doing a lot of work here, and for good reason. When ChatGPT arrived in November 2022, it turned large language models into something ordinary people could actually use. Robotics has had the research, the demos, and the funding for a long time. What it hasn’t had is the equivalent of the internet-wide dataset that made language systems take off.
That is the bottleneck Karpas is expected to focus on. Waymo and other self-driving companies built their systems over years of road miles, creating datasets that kept growing market by market. General-purpose robots do not get that kind of head start. So startups are trying to fake the missing scale with simulation, synthetic data, and foundation models that can be trained across different robot forms at once.
Nvidia has a very practical reason to care. Karpas sits at the center of the company’s startup relationships across robotics, automotive, manufacturing, mobility, smart cities, and more, which means he sees the same data problem from a lot of angles. His background also runs through Stanley Black & Decker, Intellectual Ventures, Herman Miller, iRobot, and Cirque du Soleil, with stops as an architect, manufacturing engineer, startup CEO, venture studio executive, and corporate venture capitalist.
Disrupt 2026 runs October 13-15 at Moscone West in San Francisco, where more than 10,000 tech leaders, founders, and investors are expected to show up. Nvidia is promising a lineup built around AI and real-world systems, and the company is clearly betting that robotics’ next leap will come from solving the ugly data problem, not from prettier demos.
My take — AI-written commentary, not fact-checked reporting
Robotics keeps getting treated like it’s one breakthrough away from glory, which is a very Silicon Valley habit. The dull truth is usually more useful: if the data doesn’t exist, the hype can take a number and wait outside. Nvidia is smart to press on that point instead of pretending robots just need more vibes.
Read more about this at: TechCrunch