It’s been a wild few weeks here at Automated HQ. Thankfully, we’ve got a couple of weeks to regroup before it’s back on the road. I’d meant to acknowledge when the podcast and newsletter hit the six-month mark, but things got away from me, so the fact that we just crossed that ever-important seven-month mark will have to do (6-7).
The folks at A3 whose job it is to pay attention to the metrics all tell me that things have been going great. The feedback we’ve been getting since the early days tells a similar story, so here’s a rare bit of earnestness from this perpetually jaded tech journalist thanking everyone who has supported the project over the past half-year plus.
Thanks to everyone who sought me out at GTC and Modex to talk about Automated, and all those who attended our Boston happy hour event a couple of weeks back. We’ve got a lot planned for the next several months — most of which I can’t quite talk about, but all of which is in service of delivering the best product possible.
I’ve been writing about tech for two decades and have closely covered robotics for around half of that. I spent a lot of that time wondering when this category would truly see its mainstream moment — a particular frustration after being told to focus on more important topics, like NFTs. (Remember those?)
But with mainstream coverage comes plenty of opportunities for misconception and misinformation. One of Automated’s most important goals is breaking down complex concepts in automation and physical AI in interesting and accessible ways that showcase advances, without dwelling in hyperbole. Thankfully, plenty of very smart people have our back.
Speaking of, we’ve got a ton of great podcast interviews for you in the coming months. Here’s a sample of the ones we’ve already recorded: Martial Hebert (CMU), Sergey Levine (Physical Intelligence/ U.C. Berkeley), Russ Tedrak, Daniel Rausch (Amazon), Daniela Rus (MIT), Andy Barry (Generalist), Aya Durbin (Boston Dynamics), Rick Faulk (Locus Robotics), Yoel Fink (MIT), Clara Vu, Samantha Johnson (Tatum Robotics), Matthew Johnson-Roberson (Vanderbilt University), Andrei Danescu (Dexory), and Katheryn Zealand (Skip).
The last time James Kuffner and I spoke was on the eve of world-transforming change. I interviewed TRI-AD (Toyota Research Institute – Advanced Development)’s then-CEO on March 3, 2020, at a robotics and AI event on the U.C. Berkeley campus. We set up extra Purell stands around the venue, while our corner of the world remained open for the time being.
We spoke about TRI’s plans for the Olympics in Tokyo that summer. A few weeks later, the entire event would be pushed back a year — the first such postponement in history. In the years that followed, Covid-19 tested the limits of health, communication, commerce, trade, and virtually every essential piece of human cohabitation. Technology offered the obvious solution for many, in the form of Zoom meetings and telehealth.
The moment also marked a turning point for robotics investments, as more organizations looked to automation. Companies faced difficulty meeting increased demand for e-commerce, as stay-at-home orders restricted employee movements. Even as such requirements loosened and disappeared, staffing shortages persisted, particularly in fields requiring manual labor.
Much of Kuffner’s work at the time revolved around Japan’s longstanding concerns with an aging population. In that way, TRI’s focus on automation can be seen as anticipating Covid’s fallout — even if most of the world was caught off guard by how work and care seemed to change overnight.
Kuffner changed jobs and geographies since the last time we spoke. The executive had relocated back to the States, to care for aging parents, and more recently stepped into the CTO role at Symbotic, the Massachusetts-based warehouse automation firm that made a name for itself through high-profile partnerships with grocery and retail giants like Albertsons, Target, and Walmart.
“If we're successful, if the field is successful, then it's going to change the very nature of work for people,” Russ Tedrake says. It feels like hyperbole at first blush, but it’s hard to find fault with the sentiment. "If the field is successful" is a big and important qualifier, of course, requiring tremendous time, money, and brainpower to materialize. But the promise of physical AI and robotics is one of true transformation, and all the good, bad, and unexpected that brings.
As our conversation winds down, however, the researcher mentions empathy, a word too seldom mentioned in this context.
“What are the implications for the future of factory work?” Tedrake posits. “How do we help people age in place more gracefully? It has really shaped the way I think about this. I'm spending a lot of time these days talking to labor economists and spending time with people that I think could be potentially impacted, and just trying to build my empathy muscles and build my understanding.”
A longtime MIT professor and TRI employee, Tedrake says the concept is a big piece of what ultimately pulled him into the startup world. He points to the impact generative AI has already had on the design and art world. While proponents of these technologies argue that they have served to democratize access to previously “gatekept” fields, society continues to grapple with the question of how to support those skilled artisans who have built lives around such talents.
“I can make art in a way that I couldn't a couple of years ago,” says Tedrake. “What is that going to look like in the physical world? We're going to have a lot of tasks that would have been very hard to automate before that are suddenly vibe codable almost. We'll have top craftspeople that could do magnificent things more quickly with super tools. But how does that change a culture who distributes wealth through labor? That people's sense of worth, sense of purpose often comes through the job that they've taken. We have to think very carefully about that.”
“I can't describe why,” says Sankalp Arora. “I was just always super interested in making robots.” It’s a familiar enough sentiment, surely, among our readership. What makes his expression unique, however, is where it ultimately landed him. Back home in India, Arora lacked both the financial and educational means to capitalize on said interest. Talent, perseverance, and a well-connected benefactor helped the nascent roboticist transform ambitions into his life’s work.
A few successful projects put him on the radar of Sanjiv Singh, a professor of mechanical engineering at CMU’s Robotics Institute. A loan from Arora’s father paid for a one-way ticket to Pittsburgh for an unpaid internship at the school. “[Singh] liked my work enough that he offered to fund my master’s and PhD. So that's how I got my education. I’m really, really fortunate. Really thankful for that.”
Among the projects Arora worked on during his CMU days was an autonomous helicopter, on-going research that formed the foundation of Near-Earth Autonomy, a startup cofounded by Singh in 2012 that has done key contract work for the U.S. military. Six years later, Arora founded Gather AI with fellow CMU grads, Daniel Maturana and Geetesh Dubey. In 2023, Kiva systems engineer Andrew Hoffman signed on as cofounder and CTO.
Gather was founded on the principles behind Arora’s PhD — a bid to make robots “curious,” which he defines as “an attempt to reduce uncertainty of what you think about the world. That's what our robots are doing: actively capturing data to reduce uncertainty of the warehouse environment. That's how mathematically curiosity is made.”
Before spinning the work into a standalone startup, the team that would become Gather surveyed industries in hopes of finding a profitable fit for the technology. “We knew we had the hammer of curious robots,” says Arora. “We were looking for the nail. We approached it in a very scientific fashion of doing customer studies. And before spinning out, did about 175 customer interviews.”
After nearly 15 years, Tim Cook is set to step away from Apple’s CEO role this September, settling into an executive chairman role. After more than a decade of managing the company’s supply chain, Cook suddenly found himself occupying — arguably — the industry’s largest shoes, stepping into the top spot decades before cofounder Steve Jobs’ death. His chief executive tenure largely focused on expanding Apple’s service offerings, while making large headways into China’s mobile market. More recently, however, Cook has faced criticism for the underperformance of products like the Vision Pro and Apple’s lagging AI efforts.
On September 1, another longtime Appler will take the reins. John Ternus has been with the company since 2001, three years after Cook joined. He began his time at Infinite Loop on the hardware design team, working his way up to his present role as senior vice president for hardware engineering. Ternus is a hardware guy, through and through. Before working at Apple, he worked on VR headsets. Prior to that, he studied mechanical engineering at the University of Pennsylvania. According to a January NY Times article, his senior project was “a device that allowed quadriplegics to use head motions to control a mechanical feeding arm."
While we probably shouldn’t read too much into an executive’s college days, this bit of info certainly has me wondering how active a role Ternus might play in Apple’s long-rumored robotics ambitions. The company has published a good bit of research highlighting work into home robotics that blossomed, in part, as its electric car ambitions drew to a close. Related longstanding rumors have pointed to a home hub device utilizing an iPad-like display and robot arm.
Here’s that new robot ABB was teasing a whole lot in the lead up to this week’s big show at Hanover Mess. The Zurich robotics giant is positioning the PoWa line as a kind of midway point between lightweight cobots and the bigger, heavier industrial arms the company is known for. The devices are being targeted at smaller firms looking to automate with strength and speed, while maintaining flexibility. “Cobots are growing significantly faster than traditional industrial robots, driven by demands from both small and midsized companies starting their automation journey as well as large enterprises,” says ABB Cobot head, Andrea Cassoni. “These customers are seeking higher speeds and payloads, but also greater ease of use, and compact designs. Established manufacturers want to automate heavier, fast-cycle applications, without the complexity and operational rigidity of traditional industrial robots.”
Canadian A&K Robotics Tuesday announced an $8 million CAD ($5.9 million USD) Series A for its Cruz personal mobility robots. The autonomous transport is currently focused on airport travel, with existing deployments at European and North American sites, including the company’s local Vancouver International Airport. The company describes the product as “a self-driving vehicle purpose-built for indoors.” Users enter a destination, and Cruz takes care of the rest, while avoiding fellow travelers along the way. “We’re bringing autonomy indoors,” says COO, Jessica Yip. “While others focus on roads, we’re tackling the harder problem — navigating dense, unpredictable airport crowds. Autonomous mobility is already standard in warehouses. We are bringing it into the most complex indoor environments: airports.” The system is primarily focused on passengers with mobility issues who might otherwise have difficulty making their way across massive airport terminals.
Last week, Physical Intelligence dropped π0.7, a new model it calls a “step-change.” The firm says the system can perform tasks not seen by its training data, reconstituting actions learned for other activities in ways that accomplish new ones. And yes, before you ask, that includes the ability to fold laundry without having specifically been trained on laundry folding. PI believes such outcomes represent a kind of compositional generalization observed in LLMs that has thus far not been meaningfully proven out in robotics models.
“The key to generalization for foundation models is to use broad and diverse data, which in our case includes data from many different robots, human data,” the company writes, “and even autonomous episodes collected by running various policies. Merging all these data sources naively does not lead to good results. We find that the key to using all of these data sources to attain compositional generalization is to add diverse context to the prompt: training the model with a variety of multimodal prompt structures that specify not only what the robot should do, but how it should do it. The prompt can include not just a textual description of the task, but a variety of other annotations and modalities.”
You likely didn’t need a chart to demonstrate that autonomous driving funding is up in a major way. Even so, the above blue bars from Crunchbase really drive home the industry’s second wind after a sketchy couple of years. Through Q1, this year’s figure has already tripled the entirety of 2025. Among the more notable bits in the data, however, is that we’re not looking at the same startup landgrab from a decade or so back. Rather, the big bucks are consolidated among the big names. No surprise then, that Waymo makes up the lion’s share of the $21.4 billion raised in 2026, courtesy of its $16 billion February Series D. Shield AII and Wayve round out the top three with more than $1 billion apiece.
A statement published in a recent issue of the Journal of the American Heart Association pushes for a standardized framework in a key stroke-related procedure. Mechanical thrombectomy (MT) removes blood clots from the brain. It can save lives, particularly when performed within hours of the issue’s detection. Given the level of expertise required, however, its availability is extremely limited — something medical groups are hoping can be widely addressed through robotic intervention. “While AI-assisted robotic MT is not yet ready for routine clinical use, the speed of innovation in both robotics and AI means transformative advances are likely to be just around the corner,” the paper’s senior author, Thomas Booth, told reporters. “Our work lays the foundation for that future by defining clear standards to ensure that global development and validation are well understood."
More new models for you. This one comes complete with its own custom torso. Sudo describes R1 as “a fully integrated robot system with self-developed hardware and software, powered by a manipulation-centric foundation model focused on object picking.” The goal is accurate picks of unseen objects over an extended time, relying solely on simulation data. The startup claims it was able to achieve around "98% first-attempt success and nearly 100% within two attempts, under changing lighting, dynamic backgrounds, random physical interference, and obstacle-constrained placements.”
Weeks after Ocado got Brightpick to toss a massive sheet over its LogiMAT booth in Stuttgart, the Regional Court of Hamburg has hit Elite Robots with a preliminary injunction based on a Teradyne/Universal Robots suit around software copyright. This particular bit of German legal action is unfolding at the same time as the Robotics & Assembly Automation event at the Hanover Messe. According to a Teradyne rep, Elite is taking down its booth as a result. Please shoot me an image if you're on-site.
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.
