Robotics Inflection Point
The race to build machines that can perceive, reason, and act in the physical world is entering a new phase, according to Amulya Vishwanath, a former Nvidia executive now working in venture capital. Speaking to TechCrunch's Equity, Vishwanath described physical AI as being in its "LEGO era," a shorthand for a market that is still assembling its core building blocks rather than delivering fully finished products.
Her framing captures both the excitement and the constraints shaping frontier robotics. Unlike large language models, which benefited from a relatively clean software-first scaling path, physical AI must integrate sensors, actuators, control systems, simulation, data pipelines, and safety constraints into one coherent stack. That complexity has slowed commercialization, but it has also created a wide-open field for startups trying to define the category before it hardens around a few dominant platforms.
Vishwanath's comparison suggests that the industry is still early enough for experimentation to matter more than standardization. In practical terms, that means companies are still figuring out which combinations of models, hardware, and training data can produce reliable behavior in warehouses, factories, labs, and eventually homes. The promise is enormous, but the path to scale remains uneven.
Waiting For The ChatGPT Moment
The central idea behind Vishwanath's remarks is that robotics is waiting for its own ChatGPT moment: a breakthrough that makes the technology legible to the public, useful to enterprises, and investable at scale. In generative AI, that moment arrived when a consumer-facing product translated years of research into an interface that millions could immediately understand. For physical AI, the equivalent milestone would likely be a system that can perform useful tasks across a range of environments with minimal hand-holding.
That threshold has not yet been crossed, but the ingredients are accumulating. Advances in foundation models, multimodal perception, simulation environments, and cheaper compute are making it easier to train systems that can generalize beyond narrow, pre-scripted tasks. At the same time, the robotics sector is benefiting from a broader wave of investor enthusiasm around frontier AI, even as many deployments remain pilot projects rather than full-scale rollouts.
Vishwanath's remarks also reflect a growing consensus that the next major AI platform shift may not be confined to text and images. If large language models taught the market how to think about software intelligence, physical AI could extend that logic into the material world, where machines must navigate uncertainty, friction, and real-time consequences. That leap is harder, but potentially far more economically transformative.
Why The Market Cares
Investors are watching physical AI closely because the category sits at the intersection of two powerful trends: the maturation of foundation models and the persistent labor shortages affecting logistics, manufacturing, and other operational sectors. A system that can reliably automate repetitive physical work would have immediate commercial value, especially in environments where precision, uptime, and safety are critical.
Yet the sector remains fragmented. Some startups are focused on industrial arms, others on warehouse automation, humanoid robots, or software layers that make existing machines smarter. Vishwanath's "LEGO" analogy underscores that fragmentation: the market is still assembling components, not yet converging on a single dominant architecture.
That is precisely why the moment feels consequential. Early-stage markets often look messy before they look obvious. In AI, the companies that win are frequently those that identify the right abstraction layer before the rest of the industry does. If physical AI is indeed entering its LEGO era, the next few years may determine which teams can turn modular progress into durable platforms.
For now, the message from one of the sector's closely watched voices is clear: robotics is no longer a distant promise, but it is still waiting for the breakthrough that makes it feel inevitable.