A veteran robotics researcher and founder recently expressed to me concerns around a potential “funding winter” spurred by a hypothetical humanoid bubble burst. Notably, this individual’s startup didn't specialize in humanoids, but even so, they were concerned it could "also get pulled down a bit.” They quickly clarified that even the dot com bubble burst — for all its devastating downside — had startups that survived and even thrived.
Amazon, for example, did pretty okay for itself in the long run. And at least some of the money invested in these sites advanced the larger industry. Such talk of a humanoid "bubble burst" is also purely speculative at the moment, and more a reflection of the massive sums of money that are still being lavished upon the category.
But this industry vet isn't alone in voicing such concerns. What happens to the rest of a startup ecosystem if the billions being poured into the golden child fail to deliver on a timeline deemed reasonable by those doing the pouring? Who actually feels the consequences? Who will be its Amazon and who will be its Pets.com?
I explored the topic a bit in last week’s VC feature, noting the parallels to self-driving cars, and echoing the above sentiment about developing breakthrough technologies, whether or not they arrive via the product for which they were initially intended.
Still, it’s not difficult to envision a time in the not-too-distant future when fallout from potentially disappointing humanoid returns — coupled with lord knows what state the global economy will be in — results in a broader backlash for the industry at large. All the more reason to have frank conversations around expectations, promises, and reality.
How many units constitute a pilot? What sorts of numbers are we really talking about when we talk about scale? And what is that professionally produced promotional video really showing us? Is it closer to a demo or a car commercial?
It’s my job to ask these sorts of questions. There’s also a baked-in understanding that the answer isn’t always going to be the straight forward one I'm seeking. In the case of pilots, for instance, information like this is often tightly controlled by partner companies. Generally speaking, trade secrets are secrets for a reason.
A lot of NDAs are signed in advance, for instance, limiting the level of information you’re allowed to disclose through your own channels. Often specifics are left out of press releases, and timelines are kept intentionally vague. After all, that’s the nature of piloting truly novel and disruptive technologies.
There’s a laundry list of reasons robotics startups stay away from the home: price, scalability, the wildly unstructured environment that is your teenager's floor. Safety is the one I’ve spent a lot of time thinking about of late. This is due, in part, to a kind of paradoxical relationship between human and robot at the heart of the home.
The people most in need of assistance in the home are often among society’s most vulnerable. They’re the increasing percentage of the population choosing to age in place, those living independently with disabilities, and people looking for an extra set of robot grippers to help out with children.
“You really have to take safety seriously,” says Aaron Edsinger. “There's the tactical side of it, business product side of it, and I’d say we still have a lot of work to do.”
SoftBank is set to buy ABB’s robotics division for $5.4 billion. The news comes just under six months after Zurich-based ABB Group announced plans to spin out the division. ABB Robotics has been a powerful player in the industrial robotics space, with a $2.4 billion revenue in 2024, but its parent company ultimately saw “limited synergies” between itself and the vertical. SoftBank, too, has tried its hand at robotics before. Softbank Robotics was the result of the company’s 2012 purchase of French firm, Aldebaran, a deal that resulted in the creation of the social robot, Pepper. Aldebaran ultimately returned to independence, only to declare bankruptcy earlier this year. The Japanese investment giant was also home to Boston Dynamics from 2017 to 2021, acquiring the Spot-maker from Alphabet before selling off to its current owner, Hyundai. Of this most recent deal, SoftBank chairman and CEO, Masayoshi Son, notes, “SoftBank’s next frontier is physical AI. Together with ABB Robotics, we will unite world-class technology and talent under our shared vision to fuse Artificial Super Intelligence and robotics — driving a groundbreaking evolution that will propel humanity forward.” The deal is expected to close in mid-2026.
Qualcomm announced Tuesday that it’s set to acquire Arduino, the Italian firm whose microcontroller boards have become a hobbyist mainstay over the past two decades. The move comes as Qualcomm looks to expand its presence into physical AI, in a bid to remain competitive with rival, NVIDIA. In an interview with CNBC, GM Nakul Duggal confirmed that the chipmaker views robotics as a massive potential growth category for the chipmaker — its largest behind AI. Thus far, however, Qualcomm has lacked an accessible entry point into the category for developers, prototypers, and early-stage startups. As it happens, each of those are key pieces of Arduino’s existing customer base. The deal is subject to your standard regulatory scrutiny.
We’re starting to see some key research emerge from the Amazon FAR (Frontier AI & Robotics) team around humanoid robotic control. The research division is, in part, the result of Amazon’s 2024 hiring of Covariant founders, Pieter Abbeel, Peter Chen, and Rocky Duan, along with a sizeable chunk of the startup’s staff. The initial work is OmniRetarget. In a social media post about the technology, Abbeel explains that, “High-quality re-targeting really helps the reinforcement learning.” The framework finds a Unitree robot successfully executing long motion chains with objects and obstacles of different sizes and shapes.
Remedy Robotics announced this week that its new N1 robotic surgery system has successfully completed a fully remote neurointerventional procedure — the first of its kind, according to the Bay Area firm. For those of you who would have otherwise had to Google it like I did, neurointerventional surgery refers to a form of minimally invasive procedure used to diagnose and help treat spine, neck, and head conditions. “By successfully completing these remote procedures in humans, we're not just introducing a breakthrough technology,” CEO David Bell says in a release, “we’re taking the real steps towards ensuring that every person around the world has access to the best possible endovascular care.” Remedy is also announcing $35 million in venture funding per Axios.
Inspection has long been a slam dunk robotics application. These systems — drones in particular — are really good at reaching spots we can’t and, frankly, don’t want to go. That includes cozying up to aging power plants and refineries to make sure they’re still safe for humans to be around. German firm Energy Robotics just shored up a $13.5 million Series A to add to the 32,000 hours of inspection work it has performed since its 2018 founding. Energy’s existing client roster includes Shell, BP, Repsol, BASF, Merck, and E.ON. Blue Bear Capital and Climate Investment co-led the round, which also featured Futury Capital, Hessen Capital, Kensho VC, and TADTech.
Rodney Brooks (Robust.AI)- When I want a robot reality check, I go to Rodney Brooks, who cofounded iRobot and Rethink, after spending a 25 years teaching the subject at MIT.
The Association for Advancing Automation (A3) is North America’s largest automation trade association representing more than 1,400 organizations involved in robotics, artificial intelligence, machine vision & imaging, motion control & motors, and related automation technologies.
-1.png?upscale=true&width=1120&upscale=true&name=image%20(3)-1.png)
.webp?upscale=true&width=1120&upscale=true&name=SE3+Web+Images+2+(7).webp)
.png?upscale=true&width=1120&upscale=true&name=2bb52c63-f4f6-4e7e-b34a-9e3409c1d9f8%20(1).png)
