Talk Around Town: Ultra Claims $62M Bet on Rented Warehouse Robots Could Beat Humanoids

What Folks Are Hollering About

Well, slap the mud off your boots and pull up a hay bale, because there's some loud talk comin' out of Brooklyn, of all places. Ultra, a startup the company itself describes as building dual-arm warehouse robots and leasing them under a recurring monthly model, announced on October 9, 2026 that it had pulled in $62 million across two rounds, according to the company's announcement as reported by Fortune, AI Weekly, Seedtable, and Humanoids Daily.

The company says that haul breaks down into a $50 million Series A led by Framework Ventures — with Y Combinator jumping back in — plus an earlier $12 million seed round that Y Combinator and NextView Ventures had led. Ultra also announced what it describes as a deepening tie-up with Physical Intelligence, the AI model outfit, though the precise commercial terms of that arrangement have not been confirmed by Physical Intelligence publicly.

What We Actually Know for Certain

The funding figures are about as solid as a cast-iron skillet. Fortune, AI Weekly, Seedtable, and Humanoids Daily all independently summarized the same company announcement and consistently report $62 million total — $50 million Series A plus $12 million seed — so that number is well-corroborated. One fringe source floated a figure north of $100 million, but every other outlet contradicts it, and we've kicked that one to the ditch.

Physical Intelligence is a real outfit, founded in 2024 by researchers from Google DeepMind, Stanford, and UC Berkeley, according to The Bot Scout. The company builds what it describes as general-purpose Vision-Language-Action foundation models. Its $5.6 billion Series B valuation from November 2025, led by CapitalG, is confirmed by Bloomberg and multiple specialist outlets. A reported round of roughly $11 billion that was said to be in talks as of early 2026 had not been confirmed as closed as of publication, per The Elec and Teahose.

It is also confirmed that Physical Intelligence had not begun general commercial sales as of mid-2026, with no public pricing structure released, according to Teahose and other specialist coverage. Its relationships with partners like Ultra appear to operate through individual contracts rather than any broadly available product, which means the partnership is real but its commercial weight is genuinely hard to assess from the outside.

The Self-Reported Stuff — Attributed Every Time, Y'all

Now here's where the barn door gets a little drafty. Ultra's CEO, Jon Miller Schwartz, told Fortune that the company's lease-it-by-the-month approach — customers pay an upfront integration fee and then an ongoing monthly fee for hardware and software support, the company says — has generated significant revenue and has even allowed Ultra to raise its prices. Those are some mighty fine claims, like a rooster bragging about how big his eggs are, but no independent party has verified the revenue figures, the fleet size, or the pricing trajectory.

Ultra also describes its model as lowering the capital barrier for warehouse operators who might otherwise balk at buying machines outright, according to Fortune. The company further describes the Physical Intelligence arrangement as a deepening partnership, with Ultra building the hardware and Physical Intelligence supplying the AI that, Ultra claims, lets each unit adapt to a specific customer's warehouse setup. All of that framing comes from Ultra's own communications and its CEO's statements to Fortune — not from Physical Intelligence or any independent auditor.

What Remains Murkier Than a Catfish Pond

The commercial depth of the Physical Intelligence partnership is unverified from Physical Intelligence's side. Since that company is pre-revenue and operates through individual contracts rather than a standard product, it is genuinely unclear how much of Ultra's pitch depends on a relationship that could look very different depending on who's describing it.

Beyond that, Ultra's CEO's remarks to Fortune about order volume and fleet growth are unverified self-reports. There are no customer testimonials, third-party audits, or independent revenue confirmations anywhere in the available record. The analytical pieces from AI Weekly and Black Scarab AI both add useful framing but are derivative of the Fortune original — they're not digging up new dirt, just turning over the same clod.

Analysis: The RaaS Play and the Humanoid Jab

This is analysis, not reporting, so salt accordingly. The robots-as-a-service angle — sometimes called RaaS — is a genuinely interesting structural bet. If Ultra's self-described model holds up, shifting capital expenditure off the customer's balance sheet and onto a monthly line item could meaningfully reduce friction for warehouse adoption. The question is whether the unit economics actually work at scale, and nothing in the public record confirms that they do yet.

Separately, Ultra's CEO took a swing at humanoid robots in his Fortune interview, arguing — and this is his stated opinion, not a settled finding — that practical non-humanoid robots doing real warehouse work today are delivering more immediate real-world impact than humanoids, which he described as prone to falling over and difficult to deploy reliably. That's a spicy argument to make when half the venture world is currently throwing money at humanoid startups like feed at a county fair trough. Whether the market agrees is a different question entirely, and one this funding round alone cannot answer.

Physical Intelligence's own trajectory adds a layer of complexity worth noting in any analysis. A company that is pre-revenue and structuring deals through individual contracts is a genuine partner, but it is also a partner whose own commercial model is still being figured out. Ultra's pitch leans heavily on that relationship as a differentiator — which means any wobble in Physical Intelligence's trajectory could matter a great deal to Ultra's story.