NextFin News - Physical Intelligence, the San Francisco-based startup developing a "universal brain" for robotics, is in advanced discussions to raise $1 billion in a new funding round that would propel its valuation to $11 billion. The deal, first reported by The Information on March 27, 2026, marks a doubling of the company’s valuation in less than a year, signaling a shift in venture capital appetite toward the "physical AI" layer that connects digital intelligence with mechanical action.
The proposed financing follows a $600 million Series B round led by Alphabet’s CapitalG just months prior, which valued the firm at $5.6 billion. This rapid escalation in price reflects a growing conviction among elite investors that the next frontier of generative AI lies not in chatbots, but in foundation models capable of controlling any robotic hardware—from industrial arms to humanoid assistants. Physical Intelligence, founded in 2024 by a team of alumni from Google DeepMind, OpenAI, and Stanford, has positioned itself as a software-first player, avoiding the capital-intensive pitfalls of hardware manufacturing to focus on the "operating system" of the physical world.
The $11 billion valuation is particularly striking given that Physical Intelligence has yet to report significant commercial revenue. According to Julia Hornstein of The Information, the talks involve a mix of existing backers and new sovereign wealth interest, though the final terms remain subject to change. This "pre-revenue premium" is becoming a hallmark of the 2026 AI market, where investors are betting on the scarcity of talent and the massive data moats required to solve robotic generalization. Unlike LLMs that train on the internet’s vast text, physical AI requires "robot tokens"—data from real-world interactions—which are exponentially harder to acquire.
However, the deal arrives as some corners of the market express caution. Analysts at several boutique tech research firms have noted that while the "scaling laws" for text and video are well-documented, it is not yet certain they apply equally to the messy, unpredictable physics of the real world. A $11 billion valuation for a software startup without a deployed product assumes a near-flawless execution of its "π0" (pi-zero) model across diverse environments. If the model fails to generalize beyond controlled lab settings, the company could face the same "trough of disillusionment" that hit the autonomous driving sector in the early 2020s.
The competitive landscape is also tightening. Physical Intelligence is racing against Figure, which recently secured its own $1 billion Series C at a $39 billion valuation, and Tesla’s Optimus program. While Figure and Tesla are vertically integrated, Physical Intelligence’s strategy is to be the horizontal provider for the entire industry. If successful, the $1 billion infusion will provide the necessary runway to lease massive compute clusters and hire the specialized talent needed to bridge the gap between digital reasoning and physical execution.
Explore more exclusive insights at nextfin.ai.
