This week’s edition is brought to you by Virewirx. Enable full autonomy with Virewirx’s VX60, a high-performance 60 GHz wireless system for fast and scalable robotaxi data movement.
Top Story of the Week
Lyft riders can now hail Waymo robotaxis in Nashville (link, no paywall).
Through the partnership, Lyft customers may be matched at no additional cost with a Waymo autonomous vehicle when they order a regular ride. In order for a match to happen, the desired route must start and end within Waymo’s existing service area. Users can decline a driverless ride or opt out entirely.
I’m bullish on this model. Waymo can keep serving customers who specifically want a Waymo and may be willing to wait longer or pay more for the experience, while still tapping into Lyft’s larger customer base during lower-demand periods. That feels like a win-win, and it solves a lot of the issues we’ve highlighted with the “Waymo on Uber” model.
The main downside for Lyft though is what I like to call the Trojan horse problem: a Lyft customer gets matched with a Waymo, has a great experience, and then decides to start booking directly through Waymo instead.
But I think that’s a risk Uber and Lyft are going to have to live with. It’s tough for them to compete with Waymo on the actual ride experience, so their advantage needs to come from lower prices, shorter ETAs and better reliability.
Other Stuff
Self-Driving Firm PlusAl to Go Public Via Yorkville SPAC Merger (link, no paywall).
The transaction values Santa Clara, California-based PlusAI at $800 million on an equity basis before the merger with the SPAC, Texas Ventures Acquisition III Corp… The deal is expected to provide up to about $300 million in capital, mostly from the SPAC’s trust of about $236 million. The deal also includes more than $60 million of fully committed financing.
Tesla is asking people if they want to buy and run Cybercab fleets (link).
The robotaxi interest form, which was released ahead of the company’s Cybercab event in Austin, is not definitive proof that Tesla will sell its autonomous vehicles to third-party operators. But it’s certainly an indicator of where the company’s longer-term plans lie. Tesla wants to scale and it doesn’t seem to want to do it alone.
‘Unfinished business’: Travis Kalanick revisits robotaxis with new start-up (link, no paywall).
Pony.ai and Verne Kick Off Fully Driverless Robotaxi Test Rides in Zagreb (link, no paywall). This is a nice step up from their initial launch, where they still had a safety driver onboard. And it’s a meaningful milestone now that the vehicles are operating fully driverless, a first for Europe. But the deployment still looks more like a limited-corridor robotaxi service than a true citywide deployment.
One framework I’ve been thinking about for AV launches is whether the main barrier is technological or regulatory. In Europe’s case, this increasingly looks like a regulatory challenge. We already know Pony.ai’s technology can operate driverlessly from its operations in China, so the question is less whether the system can drive and more how quickly regulators will allow that capability to be deployed more broadly.
The test rides take place along a 22-kilometer route linking Verne's headquarters and one of Zagreb's main business districts with Franjo Tuđman Airport Zagreb. The fully driverless routes will be further expanded in the next few months, with the aim to cover the whole operating area in Zagreb.
Related: INTERVIEW With Pony AI CEO James Peng: “In two years, we’ll be bigger than Waymo on fleet size” (link).
Poseidon Aerospace lands $60M ahead of first pilotless test flight (link). I’ve always been skeptical of the economics of autonomous aircraft since pilots are only a small portion of airline costs. But Poseidon’s thesis is broader: going pilotless can improve utilization, eliminate crew constraints, reduce aircraft weight by removing the cockpit and life support systems, and make it easier to operate in higher-risk environments where you may not want to put a pilot in harm’s way.
The startup is embracing autonomy and remote piloting, but it’s not bothering with the vertical takeoff and landing tech that is de rigueur among so many aviation startups. There’s also no hydrogen or electric powertrain to be found. Egret and its seaplane variant, Heron, are fixed-wing aircraft powered by regular old combustion engines.
Related: Morgan Barrett interviews Sam Rohr of JAWS, one of Poseidon Aerospace’s new investors (link).
Faster Data Off the Vehicle—Faster AI on the Road
Virewirx’s VX60 High Performance Wireless System has piqued our interest—that’s why we’re partnering with them this week to share their unique millimeter wave system built to move massive AV fleet data quickly and reliably. Reach out to Brian Lee at info@virewirx.com to learn more.
Minneapolis may require driverless vehicles to have someone in the driver’s seat (link, no paywall). Not a big surprise here. Minneapolis has a large and well-organized rideshare driver community, including many immigrant drivers, and the city and state have already been at the forefront of pushing for higher pay and stronger protections for Uber and Lyft drivers.
WeRide, Uber Win Spain’s First L4 Permit for 20-Vehicle Madrid Fleet (link).
The companies target commercial robotaxi service on the Uber app by the end of 2026, which depends on further regulatory and operational milestones the permit does not itself grant. Vehicles will carry an in-car specialist during the initial phase. Mapping, route validation, and operational readiness testing begin immediately.
Can Zoox’s ‘Toaster on Wheels’ Conquer the Self-Driving Car Capital of America? (link, no paywall).
‘One side says Tesla will kill Uber. The other says Uber owns the customer relationship and can just aggregate competing robotaxi fleets. I think both miss what changes when you remove the human driver’ (link).
‘Waymo is changing multi-stop behavior. Now when you make stops, you can only chill for 5 minutes, and then must continue’ (link). Waymo’s stop product works fine if you’re picking someone up or dropping someone off. But if you’re riding solo and want to make a quick stop, you can’t really use it that way. If you get out of the vehicle, the car leaves, and you have to request another Waymo when you’re ready to continue.
That always felt like a strange product experience to me. If I add a stop, I’d expect the same car to wait for a few minutes while I run inside somewhere. A five-minute limit seems totally reasonable, but the product should actually let solo riders use those five minutes. And I’d be happy to pay for the waiting time too. I wouldn’t expect the car to sit there for free.
Tesla Cybercab doesn’t have any manual controls — but it does have a virtual joystick (link, no paywall). This could be a good solution for first responders who need to move the vehicles out of the way although it sounds like they already have some concerns.
Related: Tesla Cybercab draws concern from Texas fire department over lack of steering wheel (link, no paywall).
‘Wayve doing testing in downtown San Jose tonight at 10:13pm’ (link).
‘Younger bro follows Big bro’ (link).
What else we’re reading/listening to
How to Get Hired at Waymo (link). Our latest piece where we put together a practical guide to finding the right role, getting noticed, networking with employees and making it through the interview process at Waymo. This was a lot of fun to write and a Waymo employee already told me “building something definitely stands out”.
Building and Scaling AV Depots with Voltera CEO Frank Reig (link). My latest podcast where we discuss what actually goes into setting up an AV depot, the potential challenges in AV depot operations, and how Voltera is building the infrastructure needed to support AV fleets at scale, among other things.
AV Events
-1 to Waymo with Dmitri Dolgov, presented by South Park Commons. Happening on Monday, September 21 in San Francisco (link).
AVs behaving badly
‘Tesla Cybercab injures passenger, and then drops them off at wrong location’ (link). Obviously not a great experience, but I’ve been surprised by how few videos we’ve seen so far of Cybercabs behaving badly. There have been some awkward drop-offs and other quirks, but given how closely these vehicles are being watched, the relative lack of major incidents caught on video feels like a mildly positive signal.
Still early and purely anecdotal though.
Shout-outs
Big thanks to TDD readers David C and Micah for referring new subscribers. If there’s someone you think would enjoy TDD, feel free to forward this email or use the referral button below, and we’ll make sure to shout out your company.
Neat Jobs
Multiple roles at Rocsys (link) via Rombout Heesterman.
Senior Software Engineer - Mapping Web at Zoox (link) via Cameron Drake.
Field Support Operations Launch Lead, AV CommOps at Uber (link) via Adriana Peth.
Staff Software Engineer, Fleet Optimization at Waymo (link) via Jamelia Quiet.
Job Moves
Christian Potthast: Woven by Toyota -> Neros Technologies (link).
Maoz Friedman: Rivian -> Bedrock Robotics (link).
Until next week :)
-Harry
“Tesla is asking others to buy and run Cybercab Fleets”
IMO - if private vendors own and run cybercab fleets, they will be “on-the-hook”, not Tesla, and thus there is even less direct incentive for Cybercabs to conform to driving laws.
…
From a recent article in PopSci (below)
“Cops issuing those tickets basically approach it the same way as they would a car with a human driver. Matthew Raifman, autonomous vehicle and traffic lead at UC Berkeley’s SafeTREC, tells Popular Science that in these cases, it’s the vehicle, not any individual behind the wheel, that’s on the hook–the same as it would be for a human-driven car.”
https://www.popsci.com/technology/what-happens-when-driverless-cars-break-the-law/
Even though the trojan horse problem is still there - Waymo likely can't take more than 20% of the rideshare market for that region. This is definitely a test period for Lyft too. If the Lyft partnership falls apart, similar to Uber, then it would be sensational