Travis is back. It was only a matter of time.
Given his old compadres Emil Michael and Shervin Pishevar are back in the news, how could he, the king pooh-ba, stay in the shadows?
To a lot of us who have been in the know, Travis Kalanick has never really gone away. He has always been doing his thing, away from the limelight which ultimately became his undoing. For eight years, he was busy building Cloud Kitchens. I heard second-hand stories of progress and stories of challenges. That’s life and that’s business. And that’s startup life.
Still, the return of Travis as a spectacle has been amazing to watch. It has been worthy of putting some popcorn, firing up the (MacBook) Neo, and sitting down and consuming it all. The return says as much about him as it does about the new hype machine, aka whatever passes for media right now.
As an entrepreneur, I love Kalanick, because he makes for a great story. He knows how to do bombast right. And he knows how to play the victim right. He has the right quotes. And he knows how to stumble over himself. In short, a media person has to love Kalanick.
So for most of the day yesterday, I watched him earn his symbolic capital. (If you missed that short essay, I would suggest you read it.) It was fun to watch him drop wisdom on a streaming video podcast. It was good to see him share his very clear, coherent and convincing vision of the physical world and how AI and robotics dovetail with that world. It stands in sharp contrast to the somewhat unclear vision of physical AI shared by academic-turned-Facebooker Yann LeCun, who recently raised over a billion dollars.
Given I have covered Travis for almost two decades, I can say he has aged, as his gray beard shows. For most of my media life, I was not the guy he gave interviews to, or scoops. For those he had more compliant reporters in media. Most of my scoops were earned, and he didn’t like that one bit.
We did enjoy an occasional drink or two, and a slice of pizza. I got to know him when he was still finding his feet with Redswoosh, a company he ultimately sold to Akamai. I was with Red Herring. I was literally in the proverbial room when he and Garrett Camp had a wild and wooly idea for an on-demand limo service at Le Web in Paris. The origin story is infused with just enough typical Silicon Valley mythology. Anyway, the point is I know the man and how he plays the game. As a founder, as a CEO and most importantly as orchestrator of media.
Given all this, and considering how much I can’t stand fluff in memos and media, I decided to do my typical steam wash of the Atoms vision manifesto. Separate the fluff from the fantastic. But first, a word about the pattern. Because it matters, and will help you understand the new manifesto.
At Uber, Kalanick consistently framed external forces such as regulators, competitors, the press, and board members as the problem. The company’s internal failures were always someone else’s move on the chess board. When Susan Fowler published her essay about sexual harassment and a culture that systematically protected high performers over the people they harmed, the initial response was to investigate the investigator. When Greyball, a tool Uber built to deceive regulators and evade law enforcement, became public, it was framed as a misunderstood safety feature.
That pattern held. The company, and by extension Travis, was always being persecuted by people who didn’t understand what it was trying to do. The mission was always too important for ordinary rules to apply.
The Atoms manifesto is pretty much the same playbook. The unnamed investor, though everyone knows who it is, who timed his power grab during a family tragedy becomes the defining event of his departure. Gone are the issues and problems at Uber. Instead what remains is the Horatio Alger-like story of a man who built something transformational, was betrayed at his most vulnerable, and is now back.
What Is Real
The strongest idea in the Atoms vision is the simplest one. Kalanick calls it “Digitizing the Physical World.” I know this because I have written about it. In my life as an investor this has been a recurring theme. The physical world is messy, chaotic, and stubbornly analog, but someone will undo this Gordian knot eventually.
The vision he outlines, sensors on streets, tracking trash trucks and traffic patterns in real time, routing around chaos before it happens, has been tried successfully, if not at scale, in Porto, Portugal almost a decade ago. It was built on a company True had backed, Veniam. The concept of smart city infrastructure, physical world data, the whole thing, was tried and worked at small scale. No one cracked how to scale it. That is still the problem. That is still the opportunity.
Same with mining automation, where companies big and small are doing self-driving automation of earth movers and trucks.
I still believe in this future. The time is now. The technology is knocking on the door. What you need is someone willing to break down the old way of doing things by sheer force of will and refusal to accept that the incumbent way is the only way. A blockhead, in the best sense of the word. If you can beat the cab cabal, one of the most entrenched, politically protected, locally organized industries in the world, then you can do things most people call impossible. That’s Kalanick in a nutshell. It is not nothing.
The second real argument is about specialized versus generalized robots. The industry is currently infatuated with humanoids. Kalanick pushes back, and he is right to. I mean you know how I feel about it. I talked at length with robotics pioneer Rodney Brook about this in my interview for Crazy Stupid Tech.
The humanoid is for bespoke, low-volume, human-designed spaces. The specialized machine is for industrial scale. This distinction matters, and not enough people covering robotics are making it clearly.
Third, the physical AI tech stack he lists, sensors, compute, models, manipulation, software, operations research, manufacturing, real estate, energy, and chemistry, is genuinely intimidating as a scope of work. His point that no single company has to own all of it but cross-stack competence is the advantage is an honest and useful framing.
Fourth, and this one is underappreciated, is land. Kalanick argues that real estate is a critical and overlooked input to physical world AI. Where do you mine the minerals? Where do you manufacture? Where do you store energy? The companies that control that land have an advantage most people are not pricing in yet. That is worth watching.
The last reason I buy this vision is because he has been on this digitizing physical world theme for more than a decade and a half. We talked about it. Often. In a column I wrote for Fast Company, I compared Uber to Google. When I asked Travis if he agreed that Uber is Google-like in that they’re both data-deterministic, he said “Uber’s task is much harder, because it is about taking bits and translating them into atoms and vice versa. Google, in other words, never has to worry about a search result getting stuck behind a trash truck. The real world is a lot more complicated.”
Now you know why the new company is called Atoms.
What Is Fluff
There is a lot. I hate this kind of bullshit, and if you have read me long enough, you know that I don’t suffer nonsense like “Valuable Unknown Truths.” The entropy-and-death framing around it is grandiosity, and lacks all substance. It is venture-speak dressed as philosophy. In plain English, what he is saying is that his company is going to look for insights others miss. Every damn startup says that in their pitch to every venture fund. Of course, you can now see this phrase show up in slide decks around the valley.
My head exploded when I read “Moore’s Law squared.” It is not a thing. The claim that the cost per unit of intelligence is dropping ninety percent per year is unverifiable and uncited. Without actual hard evidence, this is not an argument.
The Tesla and Amazon autonomous futures scenario, lights-out factories, self-driving freight, autonomous delivery, reads like a 2017 pitch deck. The vision is familiar. You know that might be good as vision, but the Chinese are already doing it. Now. Shein and others have invented the new commerce. Xiaomi and BYD have rolled out lights-out factories, and created scale like none before. Self-driving is no longer a vision, it is a feature. But still, it does sound great on TV.
And lastly, the Henry Adams quote about chaos and order is garnish at best. As I said before, Travis knows how to tell a great story with a lot of emphasis. I mean you don’t raise billions if you can’t tell a story with the forcefulness of a tornado.
What Is This Then?
Bloomberg does a good job here. City Storage Systems, the company Kalanick has been running quietly since being pushed out of Uber, is being renamed Atoms. It is going after three major verticals: food, mining, and transport robotics. That is the news.
The rest is a rebranding, a reintroduction, and an attempt to reenter the public conversation on his own terms. To me, this is a specific strategy, and it did not happen by accident. (Whoever the person is behind Travis’ comeback strategy, they were criminally underpaid.)
The vision page at atoms.co is engineered content. Not a press release. Not an interview where a reporter can push back. It is a controlled document, published direct, designed to establish the terms of the story before anyone else can. This is Sources Go Direct in action.
Sources going direct has collapsed the distance between a founder’s worldview and the public’s first impression of it. The press pickup that follows is not independent verification. It is amplification of a frame that was already set. This is symbolic capitalism at work, the accumulation of reputation and attention as a form of currency. Kalanick spent the day earning it back, in real time, in public.
The coverage of Atoms has largely worked exactly as designed. The vision is treated as the news. The rebranding is treated as a launch. The Horatio Alger story, knocked down, wandered, returned stronger, is irresistible to a media world that rewards narrative over nuance and speed over depth. You gotta fill airtime and make people buy your subscriptions.
Kalanick knows this. Uber was built, in part, on the ability to move faster than the institutions trying to contain it. The same instinct is visible here.
The real story is messier and more interesting. A complicated man, with real ideas, trying to rebuild reputation and company at the same time, using the same tools and instincts that made him famous the first time. It worked at Uber, until it didn’t.
Atoms is the next test of whether it works again.
I’ll be damned if I am not interested. And I’ll be damned if I don’t want to see what he does next. That doesn’t mean I need to gulp his MBA-speak as wisdom. I will call the game as I see it being played.
March 14, 2026. San Francisco.
Login
