A former Nvidia insider who became a venture capitalist explains why robotics has not yet reached its ChatGPT moment.

NVIDIA Robotics Leader’s Seven-Year View Frames Physical AI as Lego Era

Meta description: Amulya Vishwanath, who led NVIDIA robotics product marketing for seven years, says physical AI remains in a Lego era rather than a ChatGPT moment.

Tags: NVIDIA robotics, physical AI, Techable VC, TechCrunch Equity, foundation models

Featured image: abstract modular robotics components on a neutral background, no real people, no company logos, no text

Amulya Vishwanath spent seven years leading product marketing for NVIDIA’s robotics platforms. Today, as founder and managing partner of Techable VC, she is writing first checks in physical AI across robotics, materials science, and life sciences. Her “Lego era” description matters because it challenges the expectation that a single general foundation model will instantly make robotics feel ChatGPT-like.

Background: Seven Years Inside NVIDIA Robotics

Vishwanath’s view comes from a long run inside NVIDIA robotics. She spent seven years leading product marketing for the company’s robotics platforms, a role that required explaining robotics products to markets, customers, and builders. That background shapes how she sees the field: not as a finished industry waiting for one model, but as a system whose basic building blocks are still being rebuilt. The phrase points to a field where basic parts and safety rules still have to fit together.

Development: The Lego Era Versus a ChatGPT Moment

Vishwanath says robotics is currently in its “Lego era.” The building blocks she names are actuators, robotic hands, software, and safety standards. That framing points to assembly, iteration, and reassembly rather than a single breakthrough. It runs counter to the view that robotics, powered by a general foundation model, is about to have its “ChatGPT moment.” The reference is to late 2022, when a shockingly capable chatbot launched and changed expectations for AI. General Intuition CEO Pim de Witte told TechCrunch’s Equity podcast earlier this year that a single general foundation model would make much of today’s specialized robotics work redundant and be that moment. Vishwanath’s position shifts the question from “when will one model change everything?” to “what parts must be rebuilt before systems can combine?” It is a platform-first view, not a model-first view.

Latest: Writing Checks Across Physical AI

On TechCrunch’s Equity podcast, Dominic-Madori Davis caught up with Vishwanath to discuss where she is writing first checks as founder and managing partner of Techable VC. They also talked about what she learned as one of the first women executives to spin out of Nvidia and launch her own early-stage firm investing in physical AI across robotics, materials science, and life sciences. The episode connects her NVIDIA experience with her current role: after years of marketing robotics platforms, she is now backing founders in physical AI.

Current Stage: Building Blocks Still Being Rebuilt

The timeline moves from NVIDIA robotics marketing, to the debate over a ChatGPT moment, to Techable VC’s first checks across physical AI. The stage described in the episode remains early and structural. Actuators, robotic hands, software, and safety standards are still being rebuilt. The practical signal is that robotics progress may depend less on one dramatic model release and more on the slow, modular work of making physical systems dependable enough to be combined. The takeaway is not that breakthroughs are impossible, but that the pieces may still be unstable.