The Driverless Digest

Keep On Rolling: Electrifying, Scaling & Automating The Vehicles of Tomorrow (Curbivore 2026)

Harry Campbell — Mon, 27 Apr 2026 13:24:51 GMT

Join panelists at the just concluded Curbivore conference as they discuss what it takes to electrify, scale, and automate the vehicles of tomorrow, and the key systems supporting autonomy as deployments expand. This panel featured Aileen Zhong (Director of Government Affairs, Starship Technologies), Will Graylin (CEO & Chairman, indiGOtech), Nick Allen (Head of AV/Mobility Solutions, TaskUs), John McLean (Head of Market Development, Voltera), and Ed Walker (Senior Vice President, Hub International), and was moderated by Lloyd Lee (Reporter, Business Insider).

The conversation explores the infrastructure backbone enabling autonomy, from charging and fleet support to the human-in-the-loop systems that underpin AV safety today. It also covers key regulatory hurdles in cities, barriers to deploying sidewalk robots, and the most important metrics for early pilot programs. Panelists discuss how insurability evolves with scale, including the role of real-world versus simulated miles, how remote assistance may shift globally, and how risk and compliance frameworks change as drivers are removed. The discussion also touches on labor dynamics, the infrastructure versus scale “chicken-and-egg” challenge, public and regulatory perception, and which pieces of the ecosystem will have the biggest impact on AV growth in the coming years.

Curbivore 2026 brought together leaders across autonomy, delivery, and mobility in downtown Los Angeles for a full day of panels, firesides, exhibitions, and street food, focused on the ideas shaping the future of the curb and cities. You can check out the highlights clip here to get a feel for the event, and we’ve curated recordings of the panels, firesides, and speeches here. Also check out hundreds of our event photos here.

We’re bringing that same energy to our next Urban Autonomy Summit in Detroit on June 9. After that, we’ll be in Austin this fall and back in San Francisco in the winter for our next series of AV summits.

Chapters

(00:00) Introduction and overview

(03:45) The backbone that makes autonomy work and the infrastructure layer

(09:45) The human-in-the-loop layer behind AV safety

(11:04) Regulatory hurdles facing autonomous vehicle deployments in cities

(14:05) Barriers to deploying sidewalk robots in major cities

(15:37) The most critical metrics for AV pilot deployments

(19:03) Key requirements for autonomous fleets to ensure insurability at scale

(22:09) How physical vs. simulated miles impact AV fleet insurability

(23:23) How human-in-the-loop systems will evolve as AVs scale

(25:05) Overseas remote assistance for autonomous vehicles

(27:34) How risk and compliance dynamics change as drivers are removed

(30:34) Balancing labor concerns with scaling autonomy

(33:06) The “chicken and egg” problem of infrastructure vs. scaling

(36:20) Regulator and public perception of AVs today

(40:22) Which infrastructure pieces will most impact AV scaling, and how insurance may evolve

- Harry

Urban Autonomy in America: What’s Real & What’s Next (Curbivore 2026)

Harry Campbell — Mon, 27 Apr 2026 11:31:09 GMT

Join panelists at the just concluded Curbivore conference as they discuss how autonomy has progressed over the last few years, the challenges the industry still faces, and what it will take to accelerate progress as AV services begin to scale. This panel featured Pat Tsen (Deputy Executive Director, California Public Utilities Commission), Phillip Pierce (City Policy Lead, Zoox), Jon Miller (Chief Business Officer, Nexar), Ashwini Anburajan (CEO, Obi), and Aaron Park (Chief Business Development Officer, Robotis), and was moderated by Alan Ohnsman (Senior Editor, Forbes).

The conversation explores how AV deployments are evolving in real-world environments, including Zoox’s current rollout and safety approach, the regulatory and operational hurdles that remain, and how agencies like the CPUC are handling approvals and data reporting. It also covers early pricing dynamics and consumer perception, how companies like Nexar support AV safety as services scale, the growing role of autonomous robotics and emerging use cases, and how industry players can better align to standardize testing, improve edge case readiness, and support safe, scalable growth.

Chapters

(00:00) Introduction, and overview

(06:25) Zoox’s current AV deployment position and the challenges so far

(08:51) Steps to a more harmonized system of approval for robotaxis (Federal and state), and the CPUC’s place in this system

(12:40) How the CPUC handles data reporting for AV companies in the testing phase, and what qualifies as an autonomous vehicle deployment

(14:40) Early pricing trends, and customer perception insights in current robotaxi deployments

(21:30) Zoox’s approach to safety and how they handle edge cases

(24:20) How Nexar supports AV companies and their safety cases as services begin to scale

(30:40) How the autonomous robotics market is developing and how their use cases are evolving.

(34:58) Standardizing AV testing and validation, and how to better prepare for edge cases

(42:30) Consumer perception of AVs/pricing and how it has evolved

(46:20) Projections for the future of autonomy

- Harry

Meet Dot: DoorDash’s Accelerated Autonomous Local Commerce Solution (Curbivore 2026)

Harry Campbell — Mon, 27 Apr 2026 11:28:20 GMT

Join Harry and Harrison Shih, Senior Director at DoorDash Labs, DoorDash for a fireside chat from the recently concluded Curbivore conference. They dive into DoorDash’s push into autonomy, with a focus on their latest delivery robot, DOT, and how it fits into a broader shift toward multimodal delivery. The conversation covers everything from the timing behind automation to how cities, consumers, and logistics networks are evolving alongside it.

They also unpack where DOT fits within DoorDash’s ecosystem, how the company collaborates with cities, and what real-world use cases are emerging today versus what’s coming next.

Chapters

(00:00) Introduction

(00:40) Harrison’s background and the philosophy behind DoorDash Labs

(02:29) DOT robot specs and design philosophy

(04:00) Why now is the right moment for delivery automation

(04:59) DOT’s role in a multimodal delivery ecosystem

(07:10) How DoorDash works with cities to automate delivery

(10:15) Problems autonomy solves in delivery

(12:32) Key use cases for DOT

(14:00) Consumer behavior trends in delivery

(15:44) The “other half” of the delivery equation, alongside autonomy

(19:37) How autonomous delivery evolves in cities over time

- Harry

Tesla Launches Robotaxi Service in Dallas and Houston

Harry Campbell — Fri, 24 Apr 2026 18:19:08 GMT

This week’s edition is brought to you by TaskUs — powering safe, scalable autonomous vehicle and AI operations through human-in-the-loop expertise.

Other Stuff

Robotaxis are the new millennial lifestyle subsidy (link). Interesting article I was featured in, exploring the unique pricing window we’re seeing with robotaxis right now. They are more expensive than Uber and Lyft on average, but not always. And don’t forget that you’re sitting in a $75,000 Jaguar I-Pace plus another $120k or so of AV tech.

“From a consumer point of view, it’s definitely the golden era of AV rides, very similar to like 10 years ago with Uber and Lyft,” Harry Campbell, founder of The Driverless Digest, told Sherwood News. “We don’t know what the true cost is, but you’re definitely getting a great deal on these rides.”

My advice is to take these cheap rides while they last!

Related: ‘Obi Research Reports Tesla Robotaxi Prices Are Up 41% In San Francisco’ (link).

‘Lidar vehicle spotted in Lakeview’ (link). Apparently this is a Toyota Woven development car.

San Francisco is obsessed with the safest drivers — and ignoring the ones killing people (link, no paywall).

If San Francisco were serious about safety, the focus would not be on constraining technologies that reduce harm. It would be on aggressively addressing the dominant source of that harm: human driving behavior, weak enforcement and an underbuilt street infrastructure that fails to support pedestrians and cyclists.

Uber just disclosed an 11.52% stake in Lucid, following recent investment (h/t Sawyer Meritt).

Californians’ Embrace Of Self-Driving Rides Surges 500% In 19 Months (link). A great round-up of insights from the annual Autonomous Vehicles and the City Symposium.

Waymo self-driving taxis are coming to Detroit. Will the public accept them? (link).

To build public trust, companies must increase awareness about AV technology and develop it further, says Behram Wali, lead author of the study referenced above and assistant professor in the Department of Urban Studies and Planning at the UC San Diego School of Social Sciences.

The Human Edge in AV: Scaling Swift Emergency Response

TaskUs partners with the top autonomous vehicle (AV) and AI organizations to manage the human-critical workflows that make advanced systems safer and more reliable. When one of the leading companies deployed its fleets in new cities, they also needed to scale emergency response teams. TaskUs helped by deploying expert operators, creating 24/7 incident frameworks and implementing a unique fatigue management program. Get all the details and results in this case study.

Wayne County Airport Authority (WCAA), Unifi partnership brings autonomous wheelchairs to Detroit Metropolitan Airport’s Evans Terminal (link).

‘Take a tour of Zipline’s maintenance hub for autonomous aircraft in Dallas’ (link) via Keller Clifton.

‘Waymo love of the 280 should be studied’ (link).

‘Air traffic control for autonomous cars in parking lots—brilliant idea!’ (link).

‘Look at the table below of Waymo’s launches. Count how many times “Uber” appears after 2025’ (link). This chart and Uber’s stock price speaks for itself, investors are not buying the Uber AV narrative for now, since Waymo is the only game in town. That could change over time as Uber launches with partners like Zoox in Los Angeles and Wayve/Apollo Go in London.

‘Biggest bummer about Austin is that Waymo is only available through Uber and you can’t specify that’s what you want. Complete lottery’ (link). I don’t think Waymo regrets their partnership with Uber in Atlanta and Austin, but their hybrid model with Lyft in Nashville makes a lot more sense. Users can guarantee a Waymo by booking through the company’s app, and when demand is low, Waymo can tap into Lyft’s extensive ridehail network and Flexdrive’s fleet services.

This approach takes advantage of customers willing to pay a higher price and wait longer, while also serving as a good marketing tool for new Lyft users who are unfamiliar with AVs.

Lyft CEO, David Risher: A few evenings ago I got in my first Zoox ever with Aicha Evans (link). Could this be the foreshadowing of a future partnership?

Tesla adds ‘streaks,’ other stats to track how often drivers use Full Self-Driving software (link).

Glydways raises US$170m Series C for AV network rollout (link).

Waymo temporarily pauses San Antonio operations amid flooding, vehicle entering flooded waterway (link).

‘Avride’s autonomous fleet continues to scale and is now 200 cars and counting’ (link), via Jonah B.

Exclusive: Your delivery robot will now offer the blind real-time, on-the-ground eyes around sidewalk hazards (link, no paywall). This is pretty cool!

As Coco’s robots make food deliveries for local restaurants, they continuously log every obstacle they encounter. That data feeds into Coco’s sidewalk map, updated to the minute, and under the new partnership, it will also flow to BlindSquare. The self-voicing app converts the information into spoken alerts delivered in 26 languages, warning users roughly 10 meters before they reach a hazard.

The unit economics of autonomy will be won or lost in the Infrastructure Orchestration Layer (link), via Michael B.

Related: AV Fleet Management with Ming Maa – Moove AV CEO (link).

What else we’re reading/listening to

Understanding AI by and : Human drivers keep crashing into Waymos (link).

The AV Market Strategist by : Uber’s $10bn AV Investment, Tesla Launches Robotaxi in Dallas & Houston, Waymo Autonomous Testing in London (link).

AV Events

Over 600 people came out to Curbivore last Thursday and Friday - making it one of our biggest events yet! (link).

You can check out our highlights clip, and all the photos from the event here. We’ve curated the recordings from all the speeches, panels, and firesides here as well.

Zoox is headed to the Miami Grand Prix! Catch them at the Atlassian Williams F1 team fan zone next Wednesday to Sunday, to see their robotaxi IRL and meet some of their crew (link).

Catch up with Terawatt’s Peter Cohen on stage at the upcoming ACT Expo on May 5, 2026, where he’ll break down what goes into building a charging hub for AV fleets (link).

Let us know if you’ll be attending ACT Expo and we’ll give you a shoutout in our ‘ACT Expo Guide’ next week.

AVs/Humans behaving badly

‘Waymo stopped 8 miles from destination and told me it’s a 3 minute walk’ (link). Something similar happened to me for the first time recently: I entered a destination within Waymo’s service area and it selected a drop-off location that was a 6 minute walk away.. 🤔

‘Ambulance testing out Waymo’s reaction to it’s lights’ (link).

‘Waymos block all three lanes at a flashing red light’ (link). Looks like they’re actually lining up for a drag race.

‘Waymo support remotely allows police officer to drive car out of busy Nashville intersection’ (link).

‘Person blocks entire lane of traffic by lying on hood of Waymo in Austin’ (link). Looks like it’s time to spray the windshield wipers :)

Neat Jobs

Multiple management roles at Einride, via Isabelle Ji.

Strategic Partnerships Development Manager at Waymo (link) via Shana Downs.

Robotaxi Safety Programme Lead at Wayve (link) via Oussama Ouboulane.

Job Moves

Lauren Berger: United Talent Agency -> Waymo (link).

Apoorv Narang: Lyft -> Waymo (link).

Philipp Michel: General Motors -> Motional (link).

Shout-outs

Big thanks to TDD readers at Waymo, Kyte, and to Kate K. 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.

We’re almost at 10,000 subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us. Newsletter spots are currently sold out until July/August.

Refer a friend

Until next week :)

-Harry

Lucid Lands $750 Million in Funding, Including $200 Million from Uber

Harry Campbell — Fri, 17 Apr 2026 19:22:23 GMT

This week’s edition is brought to you by Voltera, the infrastructure partner scaling autonomous fleet deployment.

Other Stuff

Waymo opens rides to everyone in Orlando and Miami (link). Riders in both cities can now hail robotaxi rides through the Waymo app, with select Miami riders also getting access to freeway rides when they express interest in the Waymo app. Here’s the launch timeline for both cities:

Miami timeline:
- 12/5/24: Waymo announced their expansion into Miami, ahead of a future launch. They also announced Moove as their fleet partner there (link).
- 11/18/25: Began fully autonomous testing in Miami ahead of launch (link).
- 1/22/26: Waymo began welcoming their first public riders in Miami (link).
- 4/15/26: Launched full public rides in Miami, including limited freeway rides.
Orlando timeline:
- 5/27/25: Waymo announced their first road trip to Orlando (link).
- 12/11/25: They began fully autonomous testing in Orlando (link).
- 2/24/26: Began welcoming first riders in Orlando (link).
- 4/15/26: Launched rides to the general public in the city.

Wayve bags $60m from semiconductor giants to roll out self-driving tech (link).

The investment comes via an extension to the company’s Series D round, with new backing from AMD, Arm and Qualcomm Ventures. It builds on the $1.2 billion the UK-based firm raised earlier this year, which brought global automakers including Mercedes-Benz, Nissan and Stellantis onto its capital table.

Waymo rolls out new vehicles in San Francisco (link, no paywall).

Waymo plans to start offering rides in them [Ojai] to the general public “soon,” Bonelli said, but declined to say exactly when.

Related: Waymo Ojai Spotted in Truckee Doing Snow Test Drives (link).

‘You actually can get more letters on your Waymo dome. See reply’ (link).

‘An interesting effort by Brad Lander to navigate the messy politics of tech and Waymo on the East Coast, big city left’ (link). Brad Lander is a former New York City Comptroller, and a Democrat running a primary campaign against Rep. Dan Goldman in New York’s 10th Congressional District.

Why Autonomous Fleets Will Struggle in Cities (link).

Built for Autonomous Fleet Deployment

Voltera develops, owns, and operates high-capacity charging infrastructure purpose-built for autonomous vehicle fleets. By managing site selection, power coordination, construction, and operations, Voltera enables rapid deployment in dense urban markets where land, power, and permitting are the biggest constraints. As autonomy scales, infrastructure becomes the bottleneck. Voltera delivers high-performance fleet hubs that remove upfront capital and execution risk—enabling faster market entry and scalable deployment.

Please reach out here to learn more about partnering with Voltera.

BADAS 2.0: The world’s best incident prediction model, period (link).

Uber begins early test rides of Lucid Gravity robotaxis equipped with Nuro autonomy systems (link).

‘Apropos of my lottery slot for the 2028 Olympics in LA I just realized that is going to be a monster moment for Waymo. Great service all over the city for a global audience. Bet they go big’ (link). Mega events are a nightmare for Uber and Lyft drivers, with congestion, street closures, and unpredictable pickup zones already breaking the system. For Waymo, that challenge looks even harder. Pickup and dropoff coordination is one of the weakest parts of the AV experience today, and large-scale events like the Olympics only amplify that problem.

So while the Olympics could be a huge visibility moment, it may also expose some of the biggest operational gaps, especially around pickups, staging, and rider coordination at scale.

‘If Waymo gets its way, 2 million workers will be out of work’ (link). More Perfect Union tends to take a pro-worker stance, and I saw a lot of negative commentary about this video online. But after watching it, I thought the core points were fairly grounded. I don’t agree with everything, but the video raises a valid question about the impact of autonomous vehicles on human drivers and doesn’t shy away from the reality that many could lose their jobs.

Related: Lyft starts Nashville robotaxi hub for Waymo fleet (link).

Flexdrive is hiring more than 70 full-time technicians, operations managers, fleet coordinators and maintenance specialists, many of them Lyft drivers.

Lyft has always had a more ‘driver-friendly’ reputation, so it was interesting to see that many of the ~70 hires came from its existing driver base. That said, if we’re being real, autonomous vehicles are unlikely to create more jobs than they replace. The whole premise is removing the driver from the car, which inherently reduces labor demand even if new roles emerge around fleet operations and maintenance.

How Waymo and Waze are pitching in to help solve L.A.’s pothole problem (link, no paywall).

‘If you’ve ever been interested in learning about the future of AV and EV transportation - there’s an abundance of opportunities taking place right now’ (link).

Scoop: Terawatt preparing $1.5B raise for robotaxi charging depots (link - paywall).

What else we’re reading/listening to

Upstarts by : Robots! Drones! How DoorDash Is Building In Autonomous Delivery (link).

Autonomy Insiders by : Inside MOIA: Autonomous Turnkey Solution for Europe (link).

: What’s blocking Waymo in DC and how can we fix it? (link).

Analyzing Waymo’s City Distribution Using Rider-Only Miles (link). I initially framed this as “market share,” but that’s not quite accurate. “City distribution” is a better fit for what the data actually shows, how Waymo’s rider-only miles are spread across its active markets, rather than its share of total rides within those cities.

AVs/Humans behaving badly

‘Determined delivery robot in China doesn’t stop for construction’ (link).

‘My friend keeps telling me a story where he ordered a Waymo from the airport in Phoenix, with a stop along the way. He left his bag in the car while he popped into the store quickly. The car took off. Customer service couldn’t send it back.’ (link). I’ve never really understood why Waymo handles stops the way they do. If you add a stop, they effectively end the ride and swap out the car, instead of just waiting like a normal ride would. They should just let you make a stop and charge for waiting time, similar to how Uber or Lyft handle it.

Human driver in Waymo robotaxi hits pedestrian on Northeast Side, San Antonio police say (link, no paywall). Well this is a little embarrassing..

Two Waymo passengers injured in hit-and-run crash in S.F. (link, no paywall).

Days after launch, Waymo vehicles block traffic at multiple Nashville intersections (link).

AV/Humans behaving “goodly” :)

Serve Robotics: We took a recent viral moment and turned it into something you don’t see every day in out-of-home—real-time, real-world, self-aware creative (link). Check out the incident that led to the apology (link).

‘A couple of food delivery robots got an assist from some helpful humans who spotted them struggling to get over a curb in Turku, Finland’ (link).

Neat Jobs

Head of Infrastructure at Moove AV (link) via Ming Maa.

Vice President Operations (Autonomous Vehicles) at TaskUs (link) via Alex Castañeda.

Job Moves

Linda Gormly: Amazon -> Zoox (link).

Shailesh Pathak: Rivian -> Waymo (link).

Whitney Weisman: Uber -> Waymo (link).

Shout-outs

Big thanks to TDD readers 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.

We’re almost at 10,000 subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us. Newsletter spots are currently sold out until August.

Refer a friend

Until next week :)

-Harry

Analyzing Waymo’s City Distribution Using Rider-Only Miles

Harry Campbell — Tue, 14 Apr 2026 15:40:48 GMT

Every quarter, Waymo updates its Safety Impact page with the latest Rider-Only (RO)1 miles driven across its operating cities: Los Angeles, San Francisco, Phoenix, and Austin2. Here’s the latest snapshot released in March 2026:

Waymo highlights this data to benchmark its performance against human drivers. And so far, the results have been fantastic. Compared to an average human driver over the same distance in its operating cities, Waymo reports significantly better safety outcomes: 92% fewer serious injury or worse crashes, 83% fewer crashes involving airbag deployment, and 82% fewer injury-causing crashes.

Historical Rider-Only Miles Data

Waymo doesn’t publish historical breakdowns of Rider-Only miles, so to get a better sense of how things are evolving, I’ve been tracking the data by city since mid-20243:

From the chart above, we can see that total rider-only miles have scaled to ~171M per quarter, roughly 5x growth since Q3 2024.

Looking at the data on a quarterly basis below, a few trends stand out: Phoenix continues to grow steadily, while San Francisco and Los Angeles have now caught up and surpassed Phoenix on a quarterly RO miles basis.

Total Rider-Only Miles

On an absolute basis, Phoenix is still in the lead, but Los Angeles and San Francisco are catching up.

It’s worth noting that all of this data reflects Rider-Only miles, not trips or revenue. Average trip length likely varies by city, so while miles are directionally correlated with usage, they’re not a perfect proxy. Still, the trends here are telling.

City Distribution Over Time

Next, we can look at how each city’s distribution of total RO miles is evolving:

Looking at both total share and quarterly changes, a few things stand out:

Phoenix’s share is declining every quarter, which is the biggest story here and a clear sign that growth is shifting to newer markets.
Los Angeles is steadily gaining share, with consistent quarter-over-quarter increases, although the rate of change appears to be slowing.
San Francisco is mostly flat over the last year, suggesting a more mature, stable market that’s growing but not expanding its relative footprint.
Austin is ramping from a small base, showing steady gains but still early and not yet a major contributor.

Quarter over Quarter Growth

Finally, when we take a look at quarter over quarter growth, every market has grown almost every single quarter. Austin shows the highest growth rates, but off a small base. Los Angeles is growing quickly and off a much more meaningful base, while San Francisco is more variable but still in a higher-growth phase. Phoenix’s slower growth reflects a more mature, scaled market.

Put together, the data shows a clear shift in where growth is happening. Phoenix still leads in total rider-only miles, but on a quarterly basis, Los Angeles and San Francisco are now generating more new miles. Austin is bringing up the rear, but showing high growth off a small base. It will be interesting to see how these numbers change over time, and as Waymo scales and provides more data on new markets like Atlanta, Nashville and beyond.

-Harry

Miles that Waymo has driven without a human driver.

Waymo does not provide data for Atlanta yet due to limited mileage, which means the results are not yet statistically significant.

Along with the help of Redditors.

Volkswagen's MOIA to Kick Off Robotaxi Testing in Los Angeles

Harry Campbell — Fri, 10 Apr 2026 16:29:50 GMT

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.

Waymo InOut of NYC

Waymo’s Robot Car Testing Ends in NYC After Permits Expire (link).

Op-ed: With Waymo Testing Halted, We Have A Rare Chance To Get Ahead of the ‘Driverless Revolution’ (link).

Across NY, debate about the inevitability of driverless cars begins (link).

While Waymo has been at the vanguard of the driverless car push in New York, Uber – a major AV player elsewhere – has surprisingly suggested that New York should slow down to get the policy right.

There was a lot of reaction this week to Waymo being effectively kicked out of NYC, but it’s worth starting with safety.

As someone generally bullish on AVs, I’m disappointed that Waymo isn’t even being allowed to continue testing with a human safety operator behind the wheel. That feels like a pretty low-risk way to keep learning, especially in a complex environment like New York City.

That said, the safety argument is more nuanced here than in other markets. Waymo often cites data suggesting its vehicles are significantly safer than human drivers, but New York is not a typical baseline. It’s one of the only major U.S. markets where rideshare drivers are more tightly regulated with licensing, commercial insurance requirements, and drug testing. If AVs are primarily replacing professional drivers rather than the average driver, the relative safety gains may be 1 smaller than the headline numbers suggest.

Longer term, cities like New York need to find a path to embrace this technology. But it’s also fair to ask how we support human drivers through that transition, especially in a market where driving is already more professionalized and regulated.

It’s also worth noting that if safety is truly the priority, there are plenty of proven tools that cities can deploy today. Things like encouraging public transit (much safer than driving a car), congestion pricing, speed cameras, and vehicle speed limiters can have a meaningful impact right now without waiting for full autonomy to arrive. It would be nice if AV executives also pushed for these changes alongside their own solution.

My biggest takeaway is that this doesn’t have to be an either-or decision. Cities can push forward on AV testing and deployment, while also investing in policies that improve safety today and protect rideshare drivers in the near term.

Other Stuff

Singapore Gets Robotaxis as Grab, WeRide Launch Driverless Cars (link, no paywall).

The company and its partner, Chinese robotaxi operator WeRide Inc., are starting a very limited autonomous-car service along two approved routes with multiple possible stops in the city-state’s Punggol neighborhood on Wednesday. An 11-vehicle fleet is designed to ferry residents to nearby amenities and transport links.

Waymo opens robotaxi service in Nashville, partners with Lyft (link). The company will begin inviting riders on a rolling basis and initially operate in a 60-square-mile geofence. A Waymo spokesperson said they currently have “dozens of vehicles” in Nashville. Riders will first hail rides through the Waymo app, with Lyft integration coming later as the service expands. Lyft will also handle fleet operations in Nashville through its Flexdrive subsidiary.

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.

DoorDash and Wing Expand Drone Delivery Partnership to Metro Atlanta (link).

Mapping Uber’s robotaxi technology advancements (link).

Baidu Silent About Failure Of 100 Robotaxis In Wuhan (link).

Waymo is offering to help cities fix their potholes (link, no paywall). I like this idea, and it reminds me of a startup I invested in, InfraHub. AVs like Waymo are constantly collecting detailed data on road conditions, lane markings, and infrastructure. That can give cities a much better, near real-time view of where maintenance is needed instead of relying on complaints or periodic checks.

It’s a good example of the “picks and shovels” side of autonomy. Even before full deployment, this data can help cities run streets more efficiently.

Municipal officials in multiple cities where Waymo operates have reached out to the robotaxi operator for help in locating potholes on their streets, assuming that Waymo kept such data. Fortunately for them, Waymo does, and it has recently decided to launch a pilot program, along with Google’s Waze, to share its pothole data with city officials.

indiGOtech Europe’s GO Bots are back in Belgium for the second time, operating on public roads and delivering groceries in a real city environment (link).

‘You can fold down the back seats in the unsupervised Tesla robotaxi. Nice for if you want to haul something like bikes, etc. And if you leave them folded down when you leave, it’ll automatically fold them back up before the next rider’ (link).

Pittsburgh startups test robotaxi parking payment system in bid to prevent lost city revenue (link).

Autonomous vehicles (AVs) still require a human to feed a meter or pay through an app — which defeats the purpose of going fully self-driving. But last week, Meter Feeder, a parking payment platform, and Mapless AI, an AV software company, successfully completed what they say is the first machine-to-machine parking transaction in the US, just in time for Waymo’s Pittsburgh deployment.

👀Spotted my first Zoox mapping car cruising through LA the other day (link).

Curbivore 2026 is just one week away, and tickets are almost sold out. We’ve got a strong lineup of autonomy partners, including Coco, Serve, Zoox, DoorDash Labs, Uber and a bunch more, and we’re excited to show you what we’ve been putting together.

Join us on April 17th in Downtown Los Angeles for a can’t miss gathering of industry experts across autonomy, mobility, and delivery shaping the future of curbside commerce. Hurry now, grab one of the last tickets and use code Autonomy25 for an extra 25% off.

‘I Really Want My Own Private Waymo’ (link).

‘Waymo is apparently sending out merch to influencers in ~~Austin~~San Antonio’ (link).

A New Asset Class (link).

Self-driving tech is already here. It’s safer, cleaner and getting better every day. But the one thing autonomous vehicles need is reliable access to power. In other words, in order for this technology to grow, it will need infrastructure to run on. Fleet operators need a network of convenient locations to charge and operate their vehicles. These sites are at the intersection of physical and digital infrastructure, they’re different from other kinds of assets in a couple of important ways.

NTSB adds latest Waymo school bus incident to growing investigation (link).

‘Elon Musk has said that Tesla’s AI self-driving will be “10x safer than human driving”. Is that good enough?’ (link).

What else we’re reading/listening to

Search Engine by PJ Vogt: The Trial of the Driverless Car (link). Interesting look at the city of Boston’s attempt to block Waymo with a nice cameo at the end by .

Why Fully Self-Driving Cars Are Almost Impossible | The Limit (link).

Related: Check out our previous interviews with AV experts Phil Koopman and Missy Cummings who were featured in the show above.

Phil & Junko on AV Safety: Weekly News: Phil & Junko Launch Our Newsletter (link). Congrats on the launch!

Zag Talk S2E6: AVs’ Impact on Cities, Mobility & Delivery (link). Jonah and I also give a preview of what's to come at Curbivore.

DoorDash’s Autonomous Delivery Strategy with Ashu Rege (link). Our latest podcast episode features Ashu Rege, VP of DoorDash Labs at DoorDash, where he breaks down the company’s approach to autonomous delivery and how it differs from the robotaxi model. He shares the story behind DoorDash’s delivery robot DOT and explains how their Autonomous Delivery Platform (ADP) supports multiple delivery modalities. We also touch on the maturity of DoorDash’s autonomy efforts, where it makes economic sense, and how the company is balancing partnerships with in-house development as it looks to scale, among other topics.

AVs/Human behaving badly

‘Waymo drives through freshly paved road’ (link).

Self-driving car runs over Mueller Lake Park duck, raising safety concerns (link). Yes, that is a real headline.

‘waymo driving the wrong way in a whataburger drive thru’ (link).

‘When we tried to hide a fifth person in the Waymo and they discovered her mid ride’ (link).

AV behaving “goodly” :)

Zoox: Our robotaxis don’t rush, but some drivers do. When a driver decided our lane was a shortcut to a right turn, our robotaxi detected the move, yielded safely, and continued on its way (link).

Neat Jobs

Autonomous Vehicle Operations Manager: Ride Hail Operations at Moove AV (link) via Ming Maa.

Strategic Partnerships Development Manager, Utilities at Waymo (link) via Matthew Fisher.

IT Support Engineer at Wayve (link) via Emily Bragg.

Senior Software Engineer, Autonomous Vehicle Performance Metrics at Motional (link) via Gayatri K Sukumar.

Job Moves

Andrew Bridges: Senior Manager Technical Operations at DoorDash Labs (link).

Shout-outs

Big thanks to TDD reader Stefan G 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.

We’re now over 9,000+ subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us. Our newsletter spots are currently sold out until August so we will be raising prices soon.

Refer a friend

Until next week :)

-Harry

This study is 20 years old but I think it makes a good case why taxi/professional drivers may be safer than the average human driver.

DoorDash’s Autonomous Delivery Strategy with Ashu Rege

Harry Campbell — Mon, 06 Apr 2026 15:22:36 GMT

This week’s podcast is brought to you by Terawatt – purpose-built charging for autonomous vehicle fleets.

Listen to this episode on Apple Podcasts, YouTube, Spotify, and Substack. Transcripts are also available by clicking on the “transcripts” button in the top right corner of this post on the app/web version. You can also access them from the email version by clicking here.

In today’s episode, I’m speaking with Ashu Rege, Vice President of DoorDash Labs at DoorDash. Ashu shares his journey through the autonomous vehicle industry, including roles at NVIDIA and Zoox, before leading autonomy efforts within one of the largest delivery platforms in the world.

We dive into what drew him to DoorDash Labs and how the team is thinking about autonomy differently from robotaxi companies. Ashu explains the origins of DoorDash’s delivery robot DOT, the goals of the DoorDash Labs, and how their Autonomous Delivery Platform (ADP) is designed to support a wide range of delivery modalities. The conversation explores what makes autonomous delivery fundamentally different from human delivery, how mature DoorDash’s delivery solutions are today, and the different categories of autonomy DoorDash is pursuing.

We also get into the economics of autonomous delivery and the unique problems autonomy is best suited to solve in delivery. Ashu breaks down how DOT operates in the real world, including its limitations and where it stands out the most. Finally, we look ahead to what’s next for DoorDash and DoorDash Labs’ autonomy efforts over the coming year, including how the company is balancing partnerships with in‑house development.

DoorDash will also be bringing a DOT delivery vehicle to our flagship Curbivore conference on April 17 in Downtown LA, and we’ll have Harrison Shih, Senior Director of DoorDash Labs, joining me for a fireside chat.

Join industry experts across autonomy, mobility, and delivery for a can’t-miss gathering shaping the future of curbside commerce. Register now and use the code Autonomy25 to save an extra 25%.

Chapters

(00:00) Introduction to Ashu Rege
(02:47) Ashu’s background in the AV industry (Nvidia, Zoox, and DoorDash).
(04:30) What excited Ashu about joining DoorDash Labs, and its autonomy goals.
(05:43) The story behind the creation of DOT and its perks
(08:23) The goal of DoorDash Labs and what they do
(09:18) DoorDash Labs’ Autonomous Delivery Platform (ADP) explained.
(10:34) One key difference between an autonomous delivery solution and a human.
(11:52) How commercially mature are DoorDash’s autonomous delivery solutions?
(12:54) DoorDash’s autonomous delivery categories, and how they compare to each other.
(14:29) Why is now the right time to scale autonomy in delivery, and how autonomous delivery differs from robotaxis.
(18:30) How DoorDash approaches the balance between partnering and building autonomous delivery solutions in-house.
(23:31) How autonomous deliveries compare to human deliveries in cost, and the unique problem autonomous deliveries solve.
(25:20) How autonomous deliveries work using DOT, its limitations, and where it stands out.
(35:11) What to expect from DoorDash and DoorDash Labs over the next year in autonomy.
(36:52) Conclusions and final thoughts

Designing Charging Hubs for Autonomous Fleets

We are excited to partner with Terawatt, one of the leading providers of charging infrastructure. To learn more about Terawatt’s network of AV charging hubs and track record of 99%+ uptime, reach out to Logan Szidik at lszidik@terawattinfrastructure.com

Notes/Links:

You can find Ashu on LinkedIn.
DOT is DoorDash’s first in-house autonomous delivery robot. You can find more info about it here (link).
DoorDash Labs is DoorDash’s robotics and automation arm. You can find more info about them on their website (link).
- Open roles at DoorDash Labs (link).

-Harry

China Robotaxi Outage Leaves Passengers Stranded on Highways

Harry Campbell — Fri, 03 Apr 2026 16:40:34 GMT

This week’s edition is brought to you by TaskUs — powering safe, scalable autonomous vehicle and AI operations through human-in-the-loop expertise.

Other Stuff

Tesla Admits Its Robotaxis Are Sometimes Driven by Remote Humans (link, no paywall). Not sure this is that big of a deal. Remote assistance is already a core part of how most autonomous systems operate today, and even limited teleoperations at low speeds can be a practical fallback. The real question is how often it’s used and in what scenarios, not whether it exists at all.

Six of the firms insisted that their remote assistance workers, who work across the US and even, in the case of Waymo, in the Philippines, never actually drive the vehicles directly. Instead, the humans provide input that the autonomous vehicle software then decides to use or ignore. Not so for Tesla. “As a redundancy measure in rare cases … [remote assistance operators] are authorized to temporarily assume direct vehicle control as the final escalation maneuver after all other available intervention actions have been exhausted,” Karen Steakley, Tesla’s director of public policy and business development, wrote

Waymo starts robotaxi services at San Antonio International Airport (link). Waymo launched its San Antonio robotaxi service in February, though it’s still not fully available to the public yet. The company has been operating an invitation-based system that it is scaling on a rolling basis, an approach it used in Dallas, Houston, and Orlando too. Waymo said on Tuesday that it has brought “tens of thousands of people” into the San Antonio service and that it plans to make the service available to all public riders soon in the city.

GM-Backed Driverless Tech Firm Momenta Said to Confidentially File for IPO in Hong Kong (link, no paywall). Momenta builds AI-powered software for both driver-assistance features and fully driverless vehicles. The company raised $300 million from General Motors in 2021 and partnered with Uber in 2025 to deploy autonomous vehicles on its platform in international markets outside the U.S. and China.

Waymo’s skyrocketing ridership in one chart (link).

Drivers in Dire Crashes Relied Too Much on Ford’s Hands-Free Technology, NTSB Says (link, no paywall).

The Human Edge in AV: Scaling Swift Emergency Response

Robotaxies & Flying Cars Will Reinvent Real Estate (link). This is definitely an optimistic view of the future, though I’m much more bullish on autonomous vehicles reshaping suburban real estate than eVTOLs. The big piece missing from this analysis is utilization. Whether by land, air, or sea, autonomous vehicles will need high utilization, especially in the early years, to generate attractive margins, or at least not lose too much money.

Related: The Impact of Autonomous Vehicles on Real Estate with Brad Hargreaves (link).

WeRide and Uber launch driverless robotaxi service in Dubai (link). Interesting timing for a launch in Dubai, but the Middle East is increasingly looking like the second biggest autonomous vehicle hotbed after the U.S.

WeRide currently operates more than 200 robotaxis in the Middle East and has reported operational profitability in the region since 2025. The company and Uber committed in February 2026 to deploy at least 1,200 robotaxis across Dubai, Abu Dhabi and Riyadh.

Humans vs Robots: Who’s Really Replacing You? (link). There will definitely be certain trips where humans outperform robots. Airports come to mind, where some passengers want help with bags and a faster, more flexible ride. AVs tend to be slower since they follow the rules, and I’m not sure that’s what you want on an airport run, especially once you hit the infamous LAX horseshoe gridlock. The real question is whether those edge cases will make up a meaningful share of trips.

The branding risk facing new robotaxi makers (link).

“Almost everybody’s first ride in an autonomous vehicle was in a Jaguar i-Pace,” says Reilly Brennan, co-founder and partner at Trucks Venture Capital, an investor in mobility startups. “Nobody got out of a Waymo ride and said, ‘I’m going to buy a Jaguar today.’”

Production of the Jaguar I-Pace came to an end in December 2024, and at the time, Waymo was only doing around 175,000 rides per week across Phoenix and San Francisco, and just starting in Los Angeles. So you couldn’t even buy one over the past year as Waymo really began to scale.

But I think the key point here is that OEMs still need to stand out. The Toyota Prius is the most common rideshare car, but it’s not exactly aspirational. If anything, “Uber Prius” has become synonymous with a pretty mediocre experience.

That’s where I actually think companies like Rivian and Lucid have an advantage. If you get into a nice Uber Rivian or Lucid robotaxi and it feels premium, quiet, and differentiated, that can absolutely translate into consumer demand. You could see a world where the robotaxi becomes the test drive. Someone takes a few great rides and thinks, I should just buy one of these.

So instead of worrying about being “white-labeled,” OEMs might want to lean into that exposure. The real opportunity isn’t just selling fleets, it’s turning high-frequency robotaxi rides into a funnel for personal vehicle ownership, or even eventually selling personal robotaxis directly to consumers.

Elmo and Upshift partner to launch remote control car subscription service in California (link).

Upshift is reinventing car ownership for the autonomous era. Members subscribe to a car that’s there when you need it, and gone when you don’t. Our partnership with Elmo allows us to build this future today, before personal autonomous vehicles are available” said Ezra Goldman, CEO and co-founder of Upshift.

Waymo Says Robotaxi Launch in Tokyo ‘Could Be Ready in a Few Months’ (link).

California lawmaker proposes that all remote operators for autonomous vehicles in the state must be based in the U.S., possess a California driver’s license, and be limited to supervising no more than three vehicles at a time (link, see bill). This would limit Waymo’s current use of remote operators based in the Philippines for deployments in California.

We’re less than two weeks away from our flagship Curbivore conference on April 17 in Downtown Los Angeles and the agenda just went live!

We’ve got a strong AV speaker lineup including Harrison Shih (DoorDash Labs), Alan Ohnsman (Forbes), Pat Tsen (CPUC), Phillip Pierce (Zoox), Lloyd Lee (Business Insider), Jon Miller (Nexar), and Ava Yazdani (Serve Robotics), along with many more leaders across the space.

Join industry experts across autonomy, mobility, and delivery for a can’t-miss gathering shaping the future of curbside commerce. Register now and use the code Autonomy25 to save an extra 25%.

What else we're reading/listening to

Zag Talk S2E5 with AV expert Missy Cummings (link).

Freakonomics Radio: In a Driverless World, Who Loses and Who Wins? (link). Check out part one of the two part series here: Are Human Drivers Finally Obsolete? (link).

Autonomous System Safety by : A Simple Rule for Level 2++ Safety Accountability (link).

How Will London’s Driver Market Respond to Autonomous Vehicles? (link).

The Humans Powering Autonomous Vehicle Operations - Omar Zoubi, TaskUs (link). Our latest podcast episode features Omar Zoubi, VP of Autonomous Mobility & Rideshare Network Strategy at TaskUs, where he breaks down how human operators support AV fleets behind the scenes, from remote assistance and edge case handling to training and improving AI systems. He discusses how TaskUs partners across the AV ecosystem, the real-world scenarios where human intervention is still required, and how these operations scale as fleets grow. We also get into cost structures, operational complexity, and how companies think about efficiency and safety as they move toward commercialization, among other topics.

AVs behaving badly

Waymo car goes wrong way in Alamo Heights school zone, concerns parents (link). Maybe Waymo just doesn’t like kids.

A School District Tried to Help Train Waymos to Stop for School Buses. It Didn’t Work (link, no paywall).

AVs behaving “goodly” :)

FSD stops and waits for a little delivery robot to cross the street lol (link).

Neat Jobs

Autonomous Vehicle Lead, Firmware Engineering at Also. (link) via Ryan Cooling.

Senior Software Engineer, Behavior Test Engineering at Waymo (link) via Natania Antler.

Communications Intern at Zoox (link) via Marisa Wiggam.

Senior Product Designer, Driverless Operations at Waabi (link) via Ross Popoff-Walker.

Job Moves

Ankit Baid: McKinsey & Company -> Waymo (link).

Johannah Albert: Carvana -> Waymo (link).

Florian Müller: Luxoft -> Momenta (link).

Clément Chidiac: Mercari -> Wayve (link).

Shout-outs

Big thanks to TDD readers at Acute Strategy, and to Chris S. 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.

We’re now over 8,000+ subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us.

Refer a friend

Until next week :)

-Harry

How Will London’s Driver Market Respond to Autonomous Vehicles?

Tym (Self Driving Insights) — Wed, 01 Apr 2026 14:14:09 GMT

Today's post comes from Tym Syrytczyk, co-founder of the Institute for Driverless Transport, an independent research organisation studying the implications and opportunities of autonomous transport in Britain. He also writes Self Driving Insights, a biweekly newsletter covering what driverless vehicles will actually mean for the country once they arrive.

Black cabs have blockaded Trafalgar Square on multiple occasions

After launching its robotaxi service in San Francisco, Waymo took just 20 months to become the city’s second largest ride-hail provider, overtaking Lyft to capture 27% of the market. Waymo plans to enter London by September of this year. Two further providers are targeting the same market: the British startup Wayve and China’s Apollo Go have announced that they will start robotaxi trials in the city this year. The prize of London’s market share is substantial. Estimates show that over 146 million taxi and private hire trips take place in London each year, the largest market in Europe.

Naturally, this has made the incumbent ‘human’ driver market, split between black cabs, and private hire drivers anxious over the future prospects of their jobs. So how will the London driver market respond to autonomous vehicles?

As you might imagine, many incumbent drivers will oppose autonomous vehicle rollouts. The average London driver works full-time at 45 hours a week. The skills that enable the jobs don’t carry as much value outside the cab and there is no adjacent career which can give drivers equivalent autonomy to what they already have.

Additionally, we know from recent history that black cabs are willing to mobilise against threats to their incomes. Uber entered London in 2012 and operated within a legal grey area. As Uber gained market share, thousands of cabbies took to the streets, blocking central London’s traffic on several occasions.

Yet, however disruptive Uber’s arrival was, it did not eliminate the need for drivers. It introduced a new class of competitor, but one that still relied on human labour. For black cab drivers, that meant there remained, at least in principle, an adjacent market they could enter by moving into private hire, even if that came with lower status, different working conditions, and the loss of the sunk costs invested in the trade. Robotaxis present a bigger threat, and so the politics of opposition will be sharper and broader than the backlash to Uber’s entry.

This piece gives the reader the tools to understand what opposition will likely emerge and lays out the dynamics that will shape this opposition. It will dig into the regulations that enabled robotaxi providers into London, the incentives at play for each player involved, and analyse the relationships between the technology providers, the incumbent drivers, and the regulators that have the power over the emergent robotaxi market.

What does London’s taxi market look like today?

London’s taxi market is complex. It includes the recognisable licensed black cabs, officially the ‘Hackney Carriage’ and ‘private hire vehicles’ which includes the newer app-based ‘ride-hail’ services provided by firms such as Uber and Lyft, as well as licensed minicab services that can only be accessed by pre-booking. From a consumer-facing angle the line between the services is blurring over time, for instance, black cabs can now be booked through the Uber app and on Freenow by Lyft. London has no cap on the total number of black cabs or private hire vehicles.

Nevertheless, the regulatory distinction between the different types of taxi services remains sharp. Black Cab drivers have a strictly regulated profession, one can only become a black cab driver if they pass ‘the Knowledge’ a test that started in 1865 and confirms that the future-driver has a near-perfect recollection of the thousands of central London’s streets. Typically test takers require 3-4 years to study for the exam. This, alongside the black-cab licensing fees and the purchase or rent of the Hackney Carriage vehicle represents a significant upfront cost. Whilst the qualification allows for flexible self-employment and a recognised license, it doesn’t provide much in terms of transferable career value.

As part of the tradeoff for these high regulatory costs, black cabs get to enjoy certain regulatory privileges. They can pick up passengers flagging them down on the street and get exclusive rights to taxi ranks. They also are exempt from the congestion charge, a £18 to £21 (USD equivalent) daily fee for driving in central London and can use bus lanes . While it is easier to become a private hire driver they don’t receive any of these privileges, with the exception of any electric vehicle drivers who are automatically exempt from the congestion charge.

The reasons why the next taxi conflict will not look like the last

The legal foundation for robotaxi entry is the Automated Vehicles Act 2024. The process to operationalise the Act known as ‘Secondary legislation’ set the basis for trialling and licensing driverless transport services, from robotaxis to autonomous buses (”automated passenger services”), as well as personal autonomous vehicle ownership.

The government’s intent became clear on 10 June 2025, when it announced that fully driverless transport licensing regulations would be brought forward to spring 2026, nearly two years ahead of the original end-2027 timeline. The broader regulatory package enabling scaled national deployment remains on course for the second half of 2027. Such accelerations are unusual: secondary legislation typically slips rather than advances, particularly under a new government. The move was the first clear signal that Whitehall intended to actively back autonomous vehicle deployment, and the first warning shot to incumbent drivers.

The multiple disrupting entrants problem

Several firms are now targeting London's robotaxi market: Wayve in partnership with Uber, Waymo, and Baidu's Apollo Go in partnership with Uber and Lyft. This complicates the incumbents' position. During Uber's entry, cabbies could focus their lawsuits on one or two targets. Now they face simultaneous entry from several players, each with its own public affairs operation lobbying regulators. When multiple firms independently advocate similar positions, regulators tend to read this as industry consensus, lending the disruptors a collective credibility that a single entrant could not achieve.

The sovereign AI problem

The Labour government has set its stakes, perhaps surprisingly, in favour of one of the robotaxi companies. When the government announced the licensing scheme, Wayve made its flagship announcement on the exact same day. This support ties into the government’s wider political ambitions to make the UK a leader in artificial intelligence deployment, ambitions they consider Wayve central to.

The UK government’s enthusiasm for AI sovereignty means that policymakers are considering the success of domestic AI firms as favourable to British national interests. The government is closely collaborating with Wayve. Three Cabinet members celebrated the firm’s Series D raise of $1.5bn at an $8.6bn valuation, and the government’s flagship £500 million Sovereign AI venture fund will be launched at a Wayve-hosted event. As Wayve’s London deployment is its first global commercial release, it will be competing directly against the larger, better-funded Waymo. Any intervention slowing the London robotaxi market would disproportionately harm Wayve, making such measures deeply unattractive to the current government.

Chart by Daniel Abreu Marques from the AV Market Strategist publication.

When can we expect opposition?

Opposition will mobilise quickly. The CEO of Addison Lee (no paywall), a major private hire provider which holds partnerships with black cab drivers, has already called for “minimum pricing” on robotaxis citing the risk that better-funded tech firms will undercut human-driven services and jobs. The incentive to act early is clear: regulations are easier to shape before a service is established, and public scepticism of robotaxis will be highest before deployment. San Francisco data suggests this window is short, with net favourability towards autonomous vehicles rising 45 points within two years of deployment.

Novelty attracts media attention, and incumbent drivers will not want early coverage to be one-sided. Every major autonomous vehicle announcement has already drawn pushback from incumbent drivers in press coverage, and both sides know that public approval drives customers and gives policymakers cover to act in their favour.

Will the incumbents (Black cabs and ride hail drivers) coordinate between each other?

Black cabs are well-positioned to coordinate opposition across multiple channels. Their membership spans many national and industry-specific unions, and the Licensed Taxi Drivers’ Association alone represents over 10,000 of London’s 16,000 licensed black-cab drivers. Local authorities routinely treat taxi drivers as a key stakeholder group, consulting them on transport and road changes. Ride-hail drivers, by contrast, face greater coordination challenges, unionisation rates among app-based drivers are likely low, and Uber drivers’ past protests failed to disrupt service access.

As for coordination between black cabs and private hire drivers, no institution bridges the two sides. The closest attempt came from the GMB union, which in the early 2010s represented both black cab and private hire drivers, until Uber’s arrival made this untenable. The GMB’s anti-Uber stance clashed directly with its Uber driver members’ interests, and those drivers eventually broke away to form their own organisations.

Who will the opposition target? What methods will they use? And what may they demand?

Incumbent drivers can lobby a range of bodies to protect their interests: national government, Transport for London, local authorities, and on the private side, firms such as airports and train stations that operate taxi pick-up zones.

Black cab drivers have historically deployed every available method: lawsuits, protests, media campaigns and lobbying at national and local government consultations. A 2014 lawsuit against Uber over the legal definition of a taxi meter was one such battleground. Backed by established union infrastructure, they will bring these same techniques to bear against robotaxi entry.

Incumbent drivers’ demands will vary by audience. At the national level, they will push for regulations that raise the price and inconvenience of robotaxi services to keep their own offerings competitive. Addison Lee’s call for minimum robotaxi pricing is an early example. Opponents will test different arguments, gravitating toward whatever best mobilises public opinion.

Arguments likely to land with the public include a call for greater accessibility requirements: all black cabs must be wheelchair accessible by law, yet no London robotaxi trial has modified vehicles beyond a standard car. Incumbents drivers could demand equivalent requirements, raising costs for operators. Further options include mandatory insurance levies and legal challenges to TfL licensing grounds that secondary legislation was rushed. Central government’s endorsement of the technology will, however, make persuading policymakers difficult.

At the local level. Planning permission for robotaxi charging and cleaning depots are likely to become a key target: in the UK, most developments, even on private land, require community consultation. At these planning committee hearings, members of the public can sway councillors to block or modify proposals on grounds such as noise, traffic, or aesthetics giving organised opposition a ready-made lever to reduce the on-the-ground infrastructural capacity of their competition.

Conclusion

The arrival of autonomous vehicles in London will be contested, litigated, and protested in the streets, the planning committees, and the corridors of Westminster. But the politics of this disruption are structurally different from the last. Incumbent drivers face legally entrenched opponents, multiple simultaneous entrants, and a central government that has, at least in the case of Wayve, bound the prestige of its technology policy to the success of the firm.

Their strategic window is narrow. The goal will be to secure durable regulatory constraints before public opinion and political momentum shift decisively against them. San Francisco offers a cautionary precedent: even Cruise’s collapse and the subsequent public backlash against autonomous vehicles did not prompt ride-hail drivers to mobilise against Waymo. London’s incumbents will be watching the new firms’ approval ratings closely. They will need to move quickly, and together, if they are to shape the market that emerges.

- Tym

The Humans Powering Autonomous Vehicle Operations (Omar Zoubi, TaskUs)

Harry Campbell — Mon, 30 Mar 2026 15:17:09 GMT

In today’s episode, I’m speaking with Omar Zoubi, VP of Autonomous Mobility & Rideshare Network Strategy at TaskUs. Omar breaks down how TaskUs supports autonomous vehicle operators behind the scenes, from remote assistance to handling edge cases that today’s AI systems still struggle to navigate.

We get into TaskUs’ role across the AV ecosystem, including who they partner with and how their human-in-the-loop model helps fleets operate safely and scale more efficiently. Omar explains the types of real-world scenarios where AVs need intervention, how those interventions feed back into improving AI systems, and what it takes to support different types of fleets with varying operational needs.

The conversation also covers the current stage of the AV industry, including how companies like Waymo are approaching remote assistance and safety, and what challenges emerge as fleets grow. We discuss operational complexity, cost structures, and how companies think about cost per mile as they move toward commercialization.

Finally, Omar shares his perspective on where TaskUs adds the most value today, how the human-in-the-loop model will evolve over time, and what the future of the AV industry looks like as autonomy matures.

Chapters

(00:00) Introduction to Omar Zoubi and TaskUs
(03:19) TaskUs’ domain of operation
(03:30) TaskUs’ AV business model, and their partners
(04:43) What services does TaskUs provide to its AV clients?
(06:10) How does TaskUs’ assistance in edge cases help AV clients improve their AI?
(07:53) Common scenarios where AV companies might need remote assistance, and how TaskUs helps.
(09:30) Differences between supporting different AV fleets
(10:41) How Omar thinks about Waymo’s remote assistance and safety
(12:40) What stage of the AV industry are we in?
(14:19) Biggest operational challenges as AV fleets start to scale
(15:46) Common traits across operators and companies in the AV industry
(17:00) How the human-in-the-loop model will evolve as AVs mature
(19:05) How do you plan for unpredictable scenarios, like the recent SF blackout?
(21:38) How AV operational costs are distributed and cost per mile
(23:10) Where does TaskUs offer the biggest value or opportunity for AV companies?
(24:00) What does the future of the AV industry look like?
(25:43) Conclusions and final thoughts

Notes/Links:

You can find Omar Zoubi on Linkedin.
TaskUs website (link).
Learn how TaskUs supports AV operations in their case study (link).

- Harry

Waymo Hits 500,000 Weekly Rides and Over 4 Million Miles

Harry Campbell — Fri, 27 Mar 2026 18:36:23 GMT

This week’s edition is brought to you by TaskUs — powering safe, scalable autonomous vehicle and AI operations through human-in-the-loop expertise.

Other Stuff

California regulator confirms Tesla is ‘not operating an autonomous vehicle service’ (link). The quote comes from my interview with CPUC Director Pat Tsen and has been making the rounds this week. It’s also the type of objective data point I like to focus on when it comes to covering Tesla.

Related: ‘Interesting commentary on $TSLA robotaxi status in CA here’ (link) via .

Uber aims to launch Europe’s first robotaxi service with Pony AI and Verne (link, no paywall).

Under the terms of the deal, Uber will provide the ridehail network, along with the customers; Verne will manage the fleet; and Pony AI will develop the autonomous driving technology. The first service will be in Zagreb, with more European markets to follow.

Related: Pony.ai Announces Unaudited Q4 & FY 2025 Financial Results (link).

Full year robotaxi revenue: $16.6 million.
1,446 vehicles in their fleet, with a 2026 year end target of 3,000 vehicles.

WeRide also announced its financial results for Q4 and full year 2025 (link).

$21.2 million robotaxi revenue, and $29.6 million gross profit for the year 2025.
2,113 vehicles in their global fleet, with a 2,600 fleet target by 2026.
Deployed across 12 countries and 40+ cities.
Remote assistance human-to-vehicle ratio: 1:40.

The Sensor Debate: Vision, LiDAR, and the Path to Real Autonomy (link). One quote that stood out from the piece by Jiajun Zhu, Co-founder and Co-CEO of Nuro:

A common argument, often framed as a first-principles view, is that because humans drive using only their eyes, machines should do the same. While that observation is factually true, it does not follow that it is the most principled design constraint for machines. Autonomous systems are not bound by human biology. In many domains, machines achieve superior performance precisely because they use sensing and capabilities that humans do not have. Mimicking human limitations is not a requirement for building the safest or most capable system.

Uber gave up on driverless cars. Now it's paying the price (link, no paywall).

Investors have noticed, punishing Uber stock over the past few months for being a potential AV laggard. Tesla and Waymo are building their own autonomous ride services, which threaten Uber.

The Human Edge in AV: Scaling Swift Emergency Response

I tried every robotaxi in America (link) via .

Who’s driving Waymo’s self-driving cars? Sometimes, the police (link).

Waymo has relied on taxpayer-funded first responders to navigate its vehicles when they encounter issues, despite the existence of the company’s own roadside assistance team. In at least six instances identified by TechCrunch, first responders have had to take control of Waymo vehicles and move them out of traffic during emergency situations, including one in which an officer was in the middle of responding to a mass shooting.

Not exactly the best use of first responders’ time, and it feels like something that should be solved operationally, not outsourced to cities. At a minimum, Waymo should be paying for that time and hassle, especially as they scale.

It also makes you wonder why more of these situations can’t be handled remotely. Waymo has a remote assistance team, but there’s still a gap between that and true teleoperation, and this is where it shows. If they want to keep expanding, they’ll need a cleaner fallback that doesn’t involve calling in the cops when they get stuck.

‘Meet the Nuro-Lucid-Uber fleet! Now almost 100 vehicles strong’ (link).

Waymo is No. 1 on Fast Company’s Most Innovative Automotive Companies list for 2026, with Baidu in second and WeRide in ninth (link).

Why a two-seater robotaxi makes more sense than you think (link, no paywall).

Why Fully Self-Driving Cars Are Almost Impossible | The Limit (link).

I flew my drone over one of the Waymo parking lots in San Francisco (link).

What else we’re reading/listening to

Cheeky Pint by John Collison: The 20-year journey to fully autonomous cars with Dmitri Dolgov of Waymo (link).

Juan Carlos: Who Runs a 150,000 Autonomous Vehicle Fleet? (link).

The AV Crisis Playbook (link).

How the CPUC Handles AV Permits, Data, and Enforcement - Pat Tsen (link). Our latest podcast episode features Pat Tsen, Deputy Executive Director for Consumer Policy, Transportation, and Enforcement at the California Public Utilities Commission (CPUC), where she outlines the role of the CPUC and how it differs from the California Department of Motor Vehicles (DMV), walks through the process to launch an AV service in California, and explains the CPUC’s technology-agnostic approach to AV regulation, including how it handles data, reporting, and enforcement, among other topics.

Related: California Regulator Says Tesla’s ‘Robotaxis’ Are More Like a Limo in the Eyes of the Law (link).

AVs/Humans behaving badly

‘The delivery robot experiment is going great’ (link).

‘Waymo blocks restaurant lot exit in Atlanta’ (link).

‘Why do I feel sorry for a limping robot, he’s just too cute’ (link).

‘I’m not going to be pushed around by the robots’ (link).

AVs/Humans behaving “goodly” :)

‘POV: you’re a delivery robot navigating the real world 🤖 We’ll call it Dot’ (link).

Related: Stay tuned for my upcoming podcast interview with Ashu Rege, Head of DoorDash Labs. Subscribe today so you don’t miss it!

‘Waymo nails Construction area perfectly!’ (link).

‘I hope their food got delivered’ (link). Feel like you should get a $5 credit for this :)

‘Nuro seamlessly handles chaotic temporary construction zone lanes’ (link).

Neat Jobs

Multiple roles at Lyft’s AV depot in Nashville (link) via Jeremy Bird.

Multiple software engineering roles at Waymo, via Mike Curtiss.

Technical Recruiter at Wayve (link) via Adam Flockton.

Job Moves

Paul DeLong: JPD Group -> MOIA America (link).

Shout-outs

Big thanks to TDD readers at Uber, Carvolution, and to Ian A. and Carter S. 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.

We’re now over 8,000+ subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us.

Refer a friend

Until next week :)

-Harry

The AV Crisis Playbook

Harry Campbell — Wed, 25 Mar 2026 16:47:06 GMT

Today’s post comes from Chessin Gertler, a creative strategist and founder focused on the future of mobility. He is also the Founder & Editor in Chief of Freedom Machine, a post-algorithmic publication exploring human movement, and runs a multidisciplinary creative strategy practice of the same name.

Every year, roughly 40,000 people die in traffic accidents in the United States. Globally, the number exceeds one million deaths annually. Autonomous vehicles will save lives. Early deployment data indicates that the safety potential is real. But between here and that future, there is an unavoidable reality: incidents will happen in perpetuity.

Not because the technology is fundamentally flawed, but because autonomous systems operate in one of the most complex environments imaginable: dense, unpredictable human systems operating at scale. AV incidents uniquely pose the risk of becoming defining events that shape public perception, trigger regulatory action, and, in some cases, alter the trajectory of entire companies.

Which leads to a simple but underdeveloped idea: the AV companies that succeed will not just have the best technology, but the best crisis playbook.

Incidents Don’t Kill Companies. Responses Do.

In 2018, Uber’s autonomous test vehicle struck and killed a pedestrian in Tempe. The contributing factors are well documented: perception challenges, disabled emergency braking, and a distracted safety driver. Uber suspended testing, faced intense scrutiny, and ultimately exited direct AV development, selling its AV division to Aurora.

Cruise’s 2023 incident followed a different pattern. The underlying event was complex: a pedestrian was first hit by a human-driven car and then struck by a Cruise vehicle. The robotaxi attempted a pullover maneuver and dragged the pedestrian. The defining issue was not the collision alone, but how Cruise communicated it. Regulators later accused the company of failing to fully disclose key details. The California DMV suspended Cruise’s permits. Operations halted, leadership turned over, and the program was fundamentally reset.

Now consider Waymo. As the most mature and widely deployed AV system in the United States by a substantial margin, it is now experiencing scrutiny at scale: interactions with buses in Austin, a cyclist incident in Santa Monica, viral clips of awkward vehicle behavior, and more. None of these are existential events individually, but collectively they reinforce that every moment is increasingly observable, recordable, and interpretable, and that interpretation, not just performance, drives perception. Waymo may be beyond the point where a single minor incident can derail the company, but it is not beyond the consequences of cumulative perception. More importantly, it is still early in its growth trajectory. The long-term opportunity remains far larger than its current footprint, and how trust is built or eroded at this stage will shape how far that expansion can go.

The Core Problem

Autonomous vehicle companies operate in a uniquely exposed environment. Every system decision happens in public, on camera, under regulatory oversight, and within an already skeptical narrative. This means crisis response is not a downstream function, but an integral part of the product.

Yet many AV companies still treat crisis response as reactive, legal-driven, siloed from engineering, and disconnected from brand positioning. This approach does not hold up under real-world conditions.

What Goes Wrong

Across these incidents, patterns emerge.

1. Delay creates narrative loss

In both Uber and Cruise’s cases, the gap between the incident and a clear explanation allowed others to define the narrative. Media, regulators, and social platforms will always fill a vacuum. If an AV company does not explain what happened quickly and clearly, someone else will. Cruise is particularly instructive here: the company knew early that its vehicle had dragged the pedestrian, but that detail was not clearly surfaced in initial communications, creating a perception of withheld information. The lesson is simple: in autonomous mobility, partial transparency is often interpreted as concealment.

2. Over-optimization for legal risk

There is a natural instinct to limit exposure, for example by withholding information or deferring to investigation. With autonomous vehicles, that strategy often backfires. What may reduce legal risk in the long term can increase regulatory scrutiny and erode public trust. In a category still building legitimacy, responses that appear legally driven rather than responsibility-driven are often interpreted as avoidance, amplifying the consequences companies are trying to contain.

3. Misalignment with stated values

Most prominent AV companies position themselves as safety-first, transparent, and human-centered, but in moments of crisis, responses often shift toward opacity, defensiveness, and minimal disclosure. This gap is where trust breaks.

The Playbook

If incidents are inevitable, then crisis response must be designed in advance. Here is what that looks like in practice.

1. Pre-defined incident response

Every AV company should have a clear, rehearsed playbook for scenarios such as pedestrian collisions, emergency vehicle interactions, viral incidents, and regulatory inquiries. This strategy should cover not only operational response, but narrative response: what to say, how quickly to say it, through which channels, and in what format. If these decisions are being made in real time, the response is already behind.

By example, consider a pedestrian crosswalk incident. Preplanned analysis and response should activate immediately. Within hours, the company should issue an initial statement acknowledging the event, confirming known facts, and outlining next steps. This should be followed by a structured update including a clear timeline, relevant context (e.g., right-of-way, visibility conditions, actions of other actors), and, where possible, annotated video or visual reconstruction showing what the system perceived and how it responded. Simultaneously, regulators should receive a complete and proactive briefing.

Public communication should be consistent across channels, updated as new information becomes available, and framed in a way that is both technically accurate and understandable to non-experts. Critically, the tone, clarity, and level of transparency should not shift under pressure. The same communication principles that define the company’s brand in normal operation must carry through in moments of crisis. Trust is built through continuity, and any break between brand values and actual behavior in these moments is immediately visible.

2. Default to structured transparency

Speed matters, but clarity matters more. This means early acknowledgment, a clear timeline of events, visual explanation when possible, and consistent updates. Releasing raw information is not enough and is often misleading. AV companies need to explain what happened in a way that non-experts can understand.

3. Make the system legible

Autonomous systems are complex, but explanations of them cannot be. The public does not need to understand the software stack, but it does need to understand what the vehicle saw, what it decided, and why it behaved the way it did. This must be explained simply, otherwise the narrative will be lost.

Without that clarity, even technically correct behavior can be misinterpreted as failure. In many cases, the more relevant question is how a human driver would have responded under the same conditions. If an AV behaves conservatively, hesitates, or yields in a way that appears awkward but avoids risk, that context needs to be made explicit. Making system behavior legible in real time is not just an explanatory exercise; it is how companies ensure that performance translates into trust.

4. Build regulatory trust before it is needed

AV companies cannot improvise trust during a crisis. Regulators should already understand how the system works, how incidents are reviewed, and how findings are communicated. Companies that treat regulators as partners, not obstacles, navigate crises differently.

5. Stay on-brand under pressure

This principle should be evident across all of the above. Crisis response is where an AV company’s brand is most visibly tested. If a company positions itself as safety-first, transparent, and responsible, its actions in moments of failure must reflect those values in practice, not just in messaging. Companies that maintain alignment between their brand principles and their behavior during incidents will be best positioned to preserve trust and sustain long-term success.

A crisis is the moment a brand becomes real, and from that, an opportunity.

Incident Reflections

With Waymo, the gap between operational excellence and narrative control has become most visible. Based on real-world examples and the framework above, there are several areas where Waymo can refine its approach:

More consistent use of visual incident explanation. In a media environment driven by short-form video, releasing clear, annotated footage early can help anchor interpretation before it fragments across social platforms.
Tighter narrative timing. Even minor delays between an incident and a clear, authoritative explanation create space for speculation to take hold.
Greater system legibility in public communications. Not just what happened, but why the system behaved as it did, in plain language.

These are strategic decisions that shape how the technology is understood. As deployments scale, the ability to guide interpretation in real time will become as important as the ability to operate safely.

The Long Game

AV companies have the potential to save millions of lives, but the transition to autonomy will continue to be messy. Incidents will continue to happen, public reactions will follow, and trust will be tested repeatedly.

Trust is built through principled brand behavior, not performance alone. It is reinforced through proactive clarity, consistency, and transparency in moments of pressure.

In the long run, the companies that succeed in autonomous mobility will not simply engineer autonomy. They will engineer trust.

- Chessin

How the CPUC Handles AV Permits, Data, and Enforcement (Pat Tsen)

Harry Campbell — Mon, 23 Mar 2026 13:04:55 GMT

This week’s podcast is brought to you by Terawatt – purpose-built charging for autonomous vehicle fleets.

In today’s episode, I’m speaking with Shao (Pat) Tsen, Deputy Executive Director for Consumer Policy, Transportation, and Enforcement at the California Public Utilities Commission (CPUC).

We start with an overview of the CPUC and Pat’s role in it. We also break down the split between the CPUC and DMV, clarifying who regulates what, and what companies actually need to do to launch an autonomous vehicle service in California, including the different permit types required along the way.

Pat explains the CPUC’s technology-agnostic approach to AV regulation, and why Tesla’s current robotaxi deployment isn’t considered an autonomous vehicle service under CPUC jurisdiction. We then dig into what it really takes to secure an AV permit, and whether the approval process is more subjective or objective in practice.

The conversation also covers enforcement, reporting, and data collection requirements for AV companies, including new stoppage event reporting rules and what data is ultimately made public. We also get into how the CPUC approaches transparency, accountability, and the balance between innovation and public safety—and where its regulatory oversight starts to reach its limits.

Chapters

(00:00) Introduction to Shao (Pat) Tsen
(03:34) Introduction to the CPUC and their areas of regulation
(05:03) Pat’s role at the CPUC
(08:37) CPUC vs DMV: Who Regulates What?
(10:54) Steps to launching an autonomous vehicle service in California—and the different types of permits
(14:05) The CPUC’s technology-agnostic role in AV regulation
(15:05) Why Tesla’s robotaxi isn’t considered an autonomous vehicle service in California
(18:35) What it takes to get an AV permit from the CPUC
(21:35) Is the CPUC’s AV permit approval process more subjective or objective?
(24:55) Enforcement, reporting, and data collection for AV companies under the CPUC’s jurisdiction
(30:21) What autonomous vehicle companies currently have to report to the CPUC, including new stoppage event requirements
(32:45) What AV company data is made publicly available?
(37:10) How the CPUC handles new edge cases and teleoperations
(41:44) Limits of the CPUC’s oversight
(43:00) Conclusions and final thoughts

Designing Charging Hubs for Autonomous Fleets

Notes/Links:

You can find Pat on Linkedin.
Relevant links for the CPUC’s AV program
- Decision and resolution numbers which set rules or approved new authority to operate
  - D.18-05-043 aka the “Pilot Decision,” created the AV Pilot program and set requirements for participants
  - D.20-11-046 as modified by D.21-05-017 aka the “Deployment Decision,” created AV Deployment program and set the requirements for participants
  - D.24-11-002 aka the AV Data Decision, expanded AV data reporting requirements including incident reporting and “stoppage event” reporting
  - R.-25-08-013 (OIR opened August 2025) – the new AV rulemaking to update policies/processes/rules for AV passenger transportation
  - General Order (GO) 157-E: TCP Regulations
  - Resolutions:
    - Resolution TL-19144 (2023) – approved Waymo for Phase I Driverless AV Passenger Service Deployment
    - Resolution TL-19145 (2023) – approved Cruise (Phase 1 driverless deployment authorization)
- AV Program Quarterly Reporting (link).
- To file a complaint, you can fill out this form Passenger complaint form and email to consumer-affairs@cpuc.ca.gov
- To submit a public comment in the ongoing AV rulemaking, as mentioned at the 41:25 timestamp, you can access the public comment tab here R2508013

-Harry

Uber Taps Rivian to Build Robotaxis in Deal Worth Up to $1.25B

Harry Campbell — Fri, 20 Mar 2026 18:04:14 GMT

We just announced the first round of speakers for our flagship Curbivore conference on April 17th in Los Angeles; grab discounted tickets while they’re still available.

Other Stuff

Former Uber CEO (Travis Kalanick) says Waymo is ahead in the robotaxi race — and Tesla is chasing a ‘ChatGPT moment’ (link, no paywall). This comes from his recent appearance on the All-In podcast, where he shared his outlook on the current autonomy landscape:

Waymo obviously is ahead; the existence proof is there. Their issue is manufacturing and scale and urgency and fierceness,” said Kalanick…The former Uber boss said Tesla had the technical “fundamentals” needed to execute its own vision for a nationwide robotaxi service — but said that the company’s ambitious goals meant it faced a more uncertain timeline.

Related: Uber co-founder Kalanick launches Atoms in specialized robotics push (link, no paywall) that, according to its website, will operate in the food, mining, and transportation industries.

Are Driverless Cars Safe? Waymo’s C.E.O. Has Been Trying to Make the Case (link, no paywall).

We’re excited to announce our first round of speakers for Curbivore 2026, happening in Downtown LA on April 17.

Join autonomy, mobility, and delivery leaders for a can’t-miss gathering re-shaping the future of curbside commerce. Register now and use the code Autonomy25 to save an extra 25%.

Trapped! Inside a Self-Driving Car During an Anti-Robot Attack (link, no paywall).

The man punched the car’s windows and tried lifting up the vehicle. He then yelled that he wanted to kill Mr. Fulop and the other two passengers for giving money to a robot. A taxi driver would have simply driven away. But Mr. Fulop’s vehicle had no driver.

Waymo’s new safety data is in (link). Over 170M miles through Dec 2025, the Waymo Driver was involved in 13x fewer serious injury or worse crashes than human drivers in those same cities.

Bolt and Nvidia to build AI backbone for European robotaxis (link).

Nvidia adds Hyundai, BYD and other automakers to self-driving tech business (link).

Kodiak AI Scales Autonomous Driving with NVIDIA DRIVE Hyperion (link).

Rivian Aims For ‘Second Largest’ Self-Driving Fleet After Tesla, CEO Says (link).

Complaint filed over Waymo allegedly transporting minors alone (link). I drove by Hamilton High School in Los Angeles the other day right at dismissal and counted five Waymos picking up students.

Waymo’s current policy doesn’t allow unaccompanied minors, and regulators in California haven’t formally approved it yet. But in practice, it’s clearly happening anyway.

What’s interesting is how this compares to Uber and Lyft. There, the restriction on minors actually makes a lot of sense given the risks that come with a human driver. Personally, I would never let a teenage girl ride alone with a rideshare driver. But with Waymo, the calculus feels different. No driver, full trip monitoring, and a controlled system arguably make it a safer and more predictable option for kids in many cases.

My Tesla Was Driving Itself Perfectly—Until It Crashed (link, no paywall). Raffi Krikorian, former head of Uber ATG (Uber’s former self-driving division), recounts his accident while riding in his Tesla running Full Self-Driving. Good reminder of the core difference between Tesla and Waymo: Waymo takes responsibility for the ride, Tesla still puts it on the driver. If Tesla ever flips that switch and owns the liability, that’s when things get interesting.

Related: ‘FSD v14.2.2.4 Sudden Disengagement sent me into a barrier. Car is totaled’ (link). The tricky part with posts like this is we don’t actually know what happened. Without the data logs, it’s unclear whether it was a true disengagement on Tesla’s side or if the driver intervened.

Google’s Waymo Eyes South Korea Market Entry (link).

👀 Lucid Gravity spotted in LA! (link).

What else we’re reading/listening to

Platform Aeronaut by : Uber Is Quietly Winning the AV Rideshare Setup (link). I think Waymo’s model with Lyft in Nashville will actually be the future: capture first party riders directly who are willing to pay more and wait longer, and use demand aggregators like Lyft and/or Uber to fill in the gaps.

That said, near to medium term, I’d be a bit concerned about Uber’s partnership with Waymo. Waymo has already announced a dozen-plus markets beyond Austin and Atlanta without Uber, which feels telling.

The Chips Powering Autonomous Driving with Augustin Friedel (link). Our latest guest podcast episode featuring the Autonomy Insiders podcast, where sits down with Augustin Friedel to discuss the critical role of high-performance silicon in autonomous driving, exploring the evolution of System-on-a-Chip (SoC) technology, the key factors driving chip selection, and the intensifying competition among providers. You can also catch up on his earlier episode with Charlotte Eisner, Head of Commercialization & Partnerships at Bolt (link).

Tech Foundation of WeRide and Bot.Auto - Tony Han & Xiaodi Hou (link).

AV Events

Panel 8 on Day 2 of the Public Law Conference 2026, hosted by the California Lawyers Association, takes place on March 20 from 12:45–1:45 PM (link). Titled “Driving the Future – Autonomous Vehicles in California,” the panel will feature Pat Tsen, Deputy Executive Director for Consumer Policy, Transportation, and Enforcement at the California Public Utilities Commission (CPUC).

We’ll also be having Pat on the TDD podcast on Monday to discuss the CPUC’s role in AV deployments across California—subscribe now so your feed updates automatically.

Every Ride AI Attendee Gets a Free Waymo Ride - Limited Tickets Remaining (link). The event is set for April 15, 2026, at the SFJAZZ Center in San Francisco, and our Curbivore conference is a couple days later on April 17th in Los Angeles.

They’ve also lined up an impressive group of speakers, including Wayve CEO Alex Kendall, Aurora co-founder and CEO Chris Urmson, Pony.ai CFO Leo Wang, and Terawatt CEO Neha Palmer, among others. TDD readers can grab tickets with $250 off here (link).

Axios’ Driving the Future of Autonomous Vehicles, presented by Aurora (link).

AVs/Humans behaving badly

Tiktoker films video breaking a Waymo’s mirrors (link).

‘Tesla driver and passenger asleep on highway’ (link).

‘Police help man after passing out in Waymo’ (link). Honestly impressive that this guy was that intoxicated during the day.

‘Waymo wants to pick me up under a highway while blocking the only traffic lane (Orlando)’ (link).

‘A child exiting a Waymo allegedly opened the door onto an oncoming car. The driver is frustrated about the insurance process’ (link). As commenters mention, I imagine it would be best to let your insurance company handle this one.

‘Ojai stuck in the middle of the intersection while traffic is coming the other way’ (link).

‘Waymo stuck on a one way HOV lane in Houston’ (link).

Neat Jobs

AV Remote Assistance Specialist at Uber (link) via Brian Kim.

Story & Content Creation, Senior Specialist at Motional (link) via Alan Hall.

Mechanical Engineer at Waabi (link) via JD Wagner.

Senior/Staff Identity and Access Management (IAM) Engineer at Zoox (link) via Natalie B.

Here’s a full list of the jobs we’ve featured (link).

Job Moves

Douglas Murphy: Revel -> Waymo (link).

Nini Ren: AWS -> Waymo (link).

Jonathan Baines: Fleet Operations Lead at Lyft Flexdrive (link).

Shout-outs

We’re now over 8,000+ subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us.

Refer a friend

Until next week :)

-Harry

The Chips Powering Autonomous Driving with Augustin Friedel

Harry Campbell — Wed, 18 Mar 2026 13:46:57 GMT

Today, we are featuring a guest podcast from , host of Autonomy Insiders, the companion podcast to his newsletter . Daniel regularly covers the global autonomous vehicle industry with a particular focus on strategy, and we’ve long enjoyed reading and supporting each other’s work.

While the show keeps a close eye on the US market, this podcast places a special emphasis on the European and Chinese ecosystems.

In this episode, Daniel Abreu Marques sits down with Augustin Friedel, an expert in software-defined vehicles (SDV) and AI-enabled mobility. They dive into the critical role of high-performance silicon in autonomous driving, exploring the evolution of System-on-a-Chip (SoC) technology, the key factors driving chip selection, and the intensifying competition among providers. They also discuss the growing trend of OEMs developing in-house chips and the long-term impact of semiconductor shortages.

Chapters

(02:16) Introduction to SOCs and Their Importance
(04:23) Critical Role of High-Performance Chips in Autonomous Driving
(08:52) Choosing the Right Chip: Performance vs. Efficiency
(12:27) Major Players in the Chip Market for Autonomous Vehicles
(15:35) The Push for In-House Chip Development by OEMs
(18:41) Competition and Collaboration in the Chip Industry
(23:46) Market Penetration of High-Performance Chips
(25:49) Level 3 Autonomy: Challenges and Perspectives
(27:51) China’s Ambition in Semiconductor Independence
(29:19) Collaborations and Acquisitions in Chip Development
(32:22) The Wake-Up Call of the Semiconductor Shortage
(34:13) The Role of Tier 1 Suppliers in the New Ecosystem
(37:45) Understanding the Chip Supply Chain
(40:56) The Importance of Node Size in Chip Production
(44:17) Evaluating the Best Chip for Level 4 Vehicles
(46:31) Key Questions for Semiconductor Strategy

Notes/Links:

You can find Daniel on Linkedin and Twitter/X.
You can find Augustin on Linkedin.
The Autonomy Insiders Podcast is available on Spotify, Apple Podcasts, YouTube and Amazon Music.
Daniel’s Newsletter: The AV Market Strategist (link).

Zoox and Uber to Team up on Robotaxis

Harry Campbell — Fri, 13 Mar 2026 16:57:55 GMT

Other Stuff

Lucid Motors shows off robotaxi concept called ‘Lunar’ (link). Lucid unveiled a new robotaxi concept called Lunar during its investor event in New York, offering an early look at how the company might approach fully autonomous mobility. The compact, two-seat vehicle is designed without a steering wheel or pedals and is expected to share a platform with Lucid’s upcoming midsize EV lineup to reduce costs and speed development.

Red states get Waymos. Blue states get studies. (X thread, full story).

Zoox starts mapping Dallas and Phoenix for its robotaxis (link). Mapping is being done using manually driven test vehicles as the first step toward deploying its autonomous robotaxis in the two cities. They plan to start with retrofitted SUVs collecting road data before eventually testing its self-driving system and later introducing its purpose-built robotaxis. With the addition of Dallas and Phoenix, Zoox now has a presence in 10 U.S. markets, including Atlanta, Austin, Las Vegas, Los Angeles, Miami, the San Francisco Bay Area, Seattle, and Washington, D.C.

Uber, Nissan, Wayve Team Up to Offer Robotaxi Services in Tokyo (link, no paywall).

The joint initiative aims to integrate Wayve’s artificial-intelligence self-driving system into Nissan’s base vehicle, which can connect to Uber’s ride-hailing platform, they said.

Related: Qualcomm and Wayve Advance Production-Ready End-to-End AI for ADAS and Automated Driving (link). We will also dig deeper into the chip side of autonomy in next week’s podcast with Augustin Friedel.

US seeks comment on Zoox petition to deploy robotaxis without steering wheels (link).

Phoenix has lived with Waymos longer than any U.S. city. Here’s what its mayor learned (link).

WeRide and Geely Farizon to Deliver 2,000 Purpose-Built Robotaxi GXRs by 2026, Advancing Large-Scale Global Commercialization (link).

‘I’ve ridden in Waymo and Pony robotaxis. Both are really amazing. We got a look at the latest Pony cars today’ (link).

Alphabet ties a portion of Sundar Pichai’s pay to Waymo for the first time (link).

Alphabet’s board approved a $130 million Waymo BPU grant that vests based on growth in Waymo’s per-unit valuation over three years and can pay out at up to 200% of the target — pushing its potential value to $260 million. The compensation package also included $45 million BPUs for Wing, its drone delivery service.

Washington Post Editorial Board: California’s false choice on autonomous trucks (link, no paywall).

A bunch of Waymo Zeekrs awaiting retrofit at Long Beach Port (link).

Related: Waymo Factory AZ March 2026 Update (link).

What else we’re reading/listening to

Autonomy Insiders by : Inside Bolt’s Autonomous Strategy: 100K Robotaxis for Europe’s Streets (link).

New Data Shows AVs Starting to Bite Into Human Rideshare (link). Our latest blog post on Gridwise’s 2026 Autonomous Vehicle Impact Report. It’s already getting some love on Reddit too (link).

AV Fleet Management with Ming Maa - Moove AV CEO (link). In this episode, I sat down with the CEO of Waymo’s top fleet partner to discuss one of the less visible but critical parts of scaling autonomy: fleet operations. We talked about why Waymo partnered with Moove AV, how AV fleet management differs from traditional rideshare fleets, how companies decide where to locate depots, maintenance, and charging infrastructure, and why fleet operations could ultimately become one of the biggest competitive moats in the autonomous vehicle industry.

AV Events

‘Don’t miss Nuro co-founder & co-CEO Dave Ferguson’s talk on the “Technical Foundations for a Universal Autonomy Platform” at Nvidia GTC 2026 next week! Tuesday, 3/17 at 3 pm’ (link).

AVs/Humans behaving badly

‘I don’t think the people in that Waymo will ever ride it again’ (link).

‘First Al robot car: waiting for the light; Second Al robot car: bro the light is broken…’ (link).

‘Austin Waymo malfunction’ (link).

‘Angry AV just drives through accident scene while blasting upbeat music’ (link).

VIDEO: Waymo vehicle stops in between train tracks, stop arm (link).

‘Tesla FSD drives through railroad crossing gate’ (link).

Waymos in unison (link).

‘Delivery robot gets stuck behind homeless person’s tent in LA’ (link).

AVs/Humans behaving “goodly” :)

‘Waymo working masterful in Miami, Florida’ (link).

‘Faceplanted delivery robot politely asks for help’ (link).

Neat Jobs

Commercial Real Estate Transaction Manager at Moove AV (link) via Jesse DaSilva.

Technical Program Manager, Hardware Programs at Waymo (link) via Daniel Ho.

Director, Production at Zoox (link) via Zoox.

Release Manager at Wayve (link) via Ekaterina Zhivutskaya.

Here’s a full list of the jobs we’ve featured (link).

Job Moves

Patrick Ware: Lyft -> Waymo (link).

Chaitanya Vasam: Uber -> Waymo (link).

Reginaldo Almeida: Zoox -> Waymo (link).

Shout-outs

Big thanks to TDD readers at Avride, and to Karan M and Matthew C 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.

We’re now over 8,000+ subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us.

Refer a friend

Until next week :)

-Harry

New Data Shows AVs Starting to Bite Into Human Rideshare

Sergio Avedian — Wed, 11 Mar 2026 14:15:17 GMT

We covered Gridwise Analytics’ previous Autonomous Vehicle Impact Report when it was released in 2025, and the company has now published its 2026 update. The new data provides one of the clearest early looks yet at how robotaxi deployments are continuing to affect traditional rideshare markets.

The changes are still relatively small, but the data suggests that in a few AV active cities, autonomous supply may be beginning to show up in the broader ride hailing ecosystem.

AVs Are Not Just Coming – They’re Already Here

AVs are operating commercially today in cities like Atlanta, Austin, Los Angeles, Phoenix, and San Francisco. These markets provide early indicators of how autonomous services from pilot programs to limited public robotaxi services interact with existing rideshare ecosystems.

Gridwise’s report analyzes proprietary driver data from Q1 2024 through Q4 2025, comparing AV-active markets to nationwide trends in driver productivity and earnings.

The big question: Are autonomous vehicles actually impacting human rideshare drivers yet? The short answer from the data is yes, but the full story is nuanced.

Trips per Hour Are Falling Faster in AV Cities

One of the clearest signals emerging from the report is that human rideshare drivers in markets with active AV deployments are experiencing steeper declines in trips per hour than the rest of the country.

Gridwise found that in Q4 2025, trips per hour fell by approximately 5.3% in AV markets, compared with a 2.6% decline nationwide. In Los Angeles, the decline was especially pronounced, nearing a 10% year-over-year drop.

This metric is important because trips per hour are a proxy for driver productivity, essentially how much work a driver can accomplish in a given shift. A larger drop suggests that demand is being siphoned off, either by reduced trip volumes or increased competition for the remaining rides.

Earnings Are Taking a Hit in Some Cities

The productivity declines seen in AV-active markets are also showing up in driver earnings. While the national average hourly gross pay for drivers increased slightly (around +1.8%), several AV markets saw year-over-year earnings declines. Los Angeles experienced about a 3.7% drop, San Francisco saw a 1.7% decrease, and Phoenix reported a small 0.4% decline.

The link between productivity and earnings isn’t surprising: fewer trips per hour, all else equal, usually means fewer dollars in the driver’s pocket. That’s the concern for traditional drivers in markets where autonomous services are filtering into demand pools.

Not All Markets Are Equal

It would be easy to read these early trends as a uniform decline for all human drivers in AV cities, but the report and related analyses suggest a more complex picture. Some AV cities show mixed results for driver utilization: the share of time drivers spend with passengers versus waiting for rides.

In Austin and Phoenix, for example, utilization has shown modest gains despite broader trends.

This underscores an important point: AV impacts aren’t monolithic or evenly distributed.

Variables like local population trends, rideshare demand patterns, tourism cycles, and platform strategy can all tilt outcomes differently from city to city.

Why AVs Affect Rideshare Drivers

So what’s driving these shifts? There are several intertwined factors:

Increased Supply via Robots: Autonomous fleets add supply to the platform ecosystems that were once exclusively human-driven. Even if AV miles are still a small fraction of total rideshare miles, their presence is diluting demand for human drivers in specific areas.
Price Competition Pressure: According to projections cited in the Gridwise report, AV profitability at scale likely requires pricing near $1 per mile, significantly lower than average human rideshare fares, which hover closer to $3.25 per mile. That pricing dynamic could create downward pressure on prices and force platforms to push AV services into price-sensitive segments first.
Shifts in Platform Strategy: Rideshare platforms are constantly optimizing algorithms for supply, demand, and pricing. The presence of AVs adds a new data layer that can affect how these algorithms prioritize ride assignments, surge pricing, and driver incentives in certain geofenced areas.

How Fast Will AVs Overtake Human Drivers?

Despite early impacts, AV vehicles are not poised to replace human rideshare drivers tomorrow. According to S&P Global projections referenced in the report, autonomous services aren’t expected to achieve parity with human-driven rideshare markets until around 2040–2041, a timeline measured in decades, not years.

This suggests that while AVs are affecting micro-dynamics in select markets today, human drivers will remain central to rideshare services for a long time. Drivers and platforms alike should expect gradual shifts, not overnight displacement.

What This Means for Drivers

For drivers in AV-active cities, the message isn’t alarmist, but it’s a wake-up call. Here’s what matters most:

Monitor local earnings and productivity trends: Market conditions can diverge dramatically from national averages.
Watch pricing and incentive changes: Adjustments in platform economics may signal broader strategic shifts related to AV deployment.
Stay adaptable: As AV technology and platform strategies evolve, readiness to shift focus across markets could make a difference in earnings resilience.

The Gridwise report presents early evidence that autonomous tech is beginning to have real economic impacts on human rideshare drivers, but it also shows that human drivers are far from obsolete. The future of rideshare will likely be a hybrid ecosystem with humans and robots serving different segments, prices, and contexts for many years to come.

Understanding these market forces now can give drivers a strategic edge as the autonomous era continues to unfold.

My Take

The Gridwise data confirms what many industry observers have been watching in cities like Los Angeles and San Francisco: robotaxis are starting to show up in the data.

Autonomous vehicles are not replacing human drivers overnight, but they do appear to be nibbling at the edges of rideshare demand in a few early markets. A 5 to 10% drop in trips per hour might not seem huge, but it is one of the first quantitative signals that AV supply may be starting to affect the broader ride-hailing ecosystem.

At the same time, the sky is not falling. Most projections still suggest that large scale displacement of human drivers is likely decades away. What we are seeing now looks more like the early stages of a hybrid phase, where robotaxis handle simpler, highly repeatable trips in dense urban areas while human drivers continue covering everything else.

For the AV industry, the takeaway is that the transition may be slow but it is becoming measurable. The more robotaxi fleets expand in cities like Phoenix, San Francisco, Los Angeles, and Austin, the easier it will be to see how AV supply changes the dynamics of the broader ride-hailing market.

- Sergio

AV Fleet Management with Ming Maa (Moove AV CEO)

Harry Campbell — Mon, 09 Mar 2026 17:20:08 GMT

In today’s episode, I’m speaking with Ming Maa, Co-Founder and CEO of Moove AV, a global fleet provider for autonomous vehicles, and Waymo’s current fleet partner in Phoenix, Miami and London.

We discussed Ming’s time at Grab and the key rideshare lessons that ultimately inspired the creation of Moove AV. Ming explains how rideshare fleet models differ inside and outside the U.S., and what separates Moove’s traditional rideshare business (Moove.io) from Moove’s AV-focused operations.

We also dive into what it actually means to be an AV fleet partner. Ming shares why Waymo chose Moove AV, the challenges behind operating autonomous fleets, and how the company works with partners across different markets. We also talk about how Moove decides where to place its fleet depots, maintenance, and charging infrastructure, along with lessons learned from managing fleets globally.

Later in the episode, Ming breaks down the operational differences between managing autonomous fleets and human-driven fleets, including the capabilities Moove AV had to build specifically for AVs. We also explore which parts of depot operations could eventually be automated, whether there are redundancies in AV fleet operations that could be trimmed to reduce costs, and why fleet management itself could become one of the biggest competitive advantages in the AV industry.

Chapters

(00:00) Introduction to Ming Maa
(02:25) Rideshare lessons from Ming’s time at Grab, and the inspiration behind Moove AV’s creation
(06:00) Rideshare fleet models inside/outside the U.S.
(08:25) The differences between Moove.io and Moove AV
(09:50) Why did Waymo choose Moove AV as a fleet partner?
(12:43) The challenges of being an AV fleet partner
(13:47) How does Moove AV work with partners in their various markets?
(15:01) How Moove AV picks their sites for fleet maintenance/charging infrastructure
(17:30) AV fleet management lessons from different markets
(21:50) How Moove AV’s partnership impacts Waymo’s operational metrics
(23:25) What capabilities did Moove AV develop for AV fleets that are different from human-driven fleets?
(25:01) What processes in AV depot operations could be automated?
(26:07) Are there redundancies in AV fleet operations that could be trimmed to cut costs/save time?
(30:20) Does fleet management become the next moat in the AV space?
(33:17) Conclusions and final thoughts

Notes/Links:

You can find Ming Maa on Linkedin.
Moove’s website (link).
The TDD newsletter edition featuring RoboDock, mentioned at the 24:59 timestamp (link).

-Harry

Coco’s New Delivery Robots Hit the Streets

Harry Campbell — Fri, 06 Mar 2026 19:06:53 GMT

Other Stuff

Self-driving truck startup Einride raises $113M PIPE ahead of public debut (link). Been a while since I’ve seen a mobility company raise via a PIPE. The oversubscribed $113M private investment in public equity comes ahead of the company’s planned public debut through a special purpose acquisition company (SPAC) and values the Swedish autonomous and electric freight startup at a $1.35 billion pre-money valuation, down from the $1.8B initially attached to the SPAC deal. Still, there appears to be solid investor appetite since the round exceeded their earlier $100 million target.

Einride is interesting because it is positioning itself as more of an end-to-end freight operating system than just a truck manufacturer. The company already operates around 200 electric heavy duty trucks for shippers like PepsiCo, Heineken, and GE Appliances across Europe, the U.S., and the UAE, while also developing its cab-less autonomous “eBot” trucks for the future. In addition to designing autonomous trucks and software, Einride also builds and manages charging infrastructure and runs its own fleet management and remote operations platform.

NTSB preliminary investigation into Waymo collision with child (link). One detail that stands out from the diagram is that the pedestrian appears to have entered the lane from the Waymo’s left but was struck near the vehicle’s front-right headlight. Hard to draw conclusions without the telemetry, but I still think it would be helpful if Waymo released the video and sensor data to show when the child was first detected and whether braking began as soon as the pedestrian became visible.

Bank of America upgrades Tesla, expecting it to “quickly become a leader in robotaxi services” (link).

Supervisors grill Waymo about 1,500 stalled cars during December blackout (link, no paywall).

Waymo’s fleet stopped 1,593 times for two minutes or more during the outage, according to company data shared with the California Public Utilities Commission last month. Waymo had to send staffers or tow trucks to retrieve 64 of its Jaguar I-PACE electric cars during the blackout. In two cases, city first responders had to get behind the wheel.

It’s still unclear to me what actually caused the 1,500+ stalls during the blackout after reading about the hearing. One thing that did stand out though: Waymo expects first responders to move stranded robotaxis in some situations, which city officials said effectively turns them into roadside assistance for the fleet. Not exactly a great look.

May Mobility’s Expanding Autonomous Footprint in Japan (link).

Jefferies analysts’ conclusions: “AVs still fall short on what matters most for adoption…. the interplay of price and ETA. However, with longer trip times and similar/higher prices (ex-Tesla in Austin), we think AVs have initially relied on hype to help drive adoption…” (link).

Robot turtles: Waymo’s next-gen cars could be buggy Hyundai EVs (link, no paywall).

Since the Ioniq 5 launched in 2021, it’s been showered with accolades…There’s just one problem: Owners — including me — say their Ioniq 5s keep failing. Drivers point to the integrated charging control unit, or ICCU, an electrical component that converts high-voltage charge to low-voltage charge. When it breaks, seemingly at random, cars can experience charging failures and power loss on the road, leaving them undriveable.

Ryan Transportation Partners with Bot Auto for Full Driverless Operations in Texas (link).

Tesla touts California robotaxis but does nothing to get permits (link).

Bryant Walker Smith, a University of South Carolina law professor and autonomous-driving expert who has consulted for the California DMV, said Tesla is implying that “they are ready and regulators are not,” while the reality is that “regulators are ready, and they are not.”

Waymo just expanded to Miami Beach and other ocean towns (link). Waymo’s service zone now covers 100 square miles, although there is still a waitlist for users who want to try the service.

You can actually buy this car with no steering wheel (link).

Waymo could expand to Virginia under new driverless car bill (link).

Pony.ai Achieves Gen-7 Robotaxi Unit Economics Breakeven in Shenzhen, Strengthening Path to Scalable Commercialization (link).

“The latest achievement in Shenzhen marks another important validation of our technology readiness and commercialization strategy,” said Dr. James Peng, Founder and Chief Executive Officer of Pony.ai. “Reaching unit economics breakeven in two Tier-1 Chinese cities demonstrates that autonomous mobility is not only technically viable, but also economically sustainable at scale.”

Harbinger acquires autonomous driving company Phantom AI (link).

‘I went home for lunch and now I’m stuck at home (link).’ Maybe he should try calling an Uber?

What else we’re reading/listening to

Compound Interest from Semafor Business: Dara Khosrowshahi on robotaxi ‘mission control’ and Uber’s next billion-dollar business (link).

Inside Waymo’s Remote Assistance Program with Dr. Missy Cummings (link). My latest podcast episode where we focused on one of the most misunderstood pieces of the AV stack: remote operations. We broke down the difference between remote assistance and true teleoperation, why latency matters more than most people realize, common misconceptions around Waymo’s remote assistance model, and the safety implications of offshore remote operators, among many other topics. Here’s one of Dr. Cummings’ spicier takes from the episode (link), and it’s getting some love on Reddit.

Watch out for Monday’s episode of the TDD podcast featuring Ming Maa, Co-Founder and CEO of Moove AV, where we discuss what it takes to manage AV fleets - you won’t want to miss it. Subscribe now so your feed automatically updates.

AV Events

Zoox is showing off its robotaxi at The LINE Austin. From March 13 to 15, visitors can see the vehicle up close and sign up for their Austin waitlist (link).

The National Highway Traffic Safety Administration (NHTSA) is hosting a National AV Safety Forum on March 10, 2026 at the USDOT Headquarters’ West Atrium in Washington, DC (link).

AVs/Humans behaving badly

‘Waymo stand-off’ (link). The third Waymo...lol.

NTSB: Human error prompted Waymo to pass stopped school bus in Austin (link). I’m liking these NTSB diagrams and descriptions. They are helpful for understanding the setup, especially since Waymo is not releasing video or sensor data.

In this situation, the Waymo vehicle initially did the right thing and stopped for the bus, but after contacting remote assistance it was incorrectly told the signals were not active and proceeded through, which resulted in a clear violation of Texas law.

But I’m not sure I blame Waymo entirely here. This looks like a pretty terrible location for a bus stop, right next to a busy multi lane 35 mph intersection. Technically, traffic in both directions has to stop even if they have a green light, which is not very intuitive. I would never teach any of my three kids to cross a street like this without a walk signal, so it seems odd to require traffic on the opposite side to stop for a bus here. The fact that several human drivers also passed the bus suggests the scenario itself may be confusing, not just the AV behavior.

‘Over the past couple of months I’ve been evaluating Rideshare use from both a travel mom and Child Passenger Safety perspective.’ (link).

‘A Waymo vehicle blocked traffic as first responders raced to the Austin bar mass shooting, where three were killed and 14 injured.’ (link). Quite unfortunate incident, but the county EMS chief said the errant Waymo had no impact on patient outcomes thankfully (link).

‘So, I almost got killed by a robot last week…’ (link).

‘This driverless Waymo car got so confused’ (link).

‘Blocking a bike lane’ (link).

‘Waymo’s Miami Emergency Protocol Failure Nearly Got Me Hit on MacArthur Causeway’ (link).

‘These Waymos are getting aggressive’ (link).

AVs behaving “goodly” :)

Nuro: Watch an autonomous passenger pick-up/drop-off maneuver in San Francisco’s busy North Beach neighborhood (link).

Waymo driving in snow (link).

Neat Jobs

Software Engineer at Einride (link) via Jonas Hatzenbühler.

AV Fleet Operations Associate (Day Shift) at Lyft (link) via Amber Lynn Stoppel.

Product Manager, In-Car Experience at Waymo (link) via Tyler Fox.

Engineering Manager, Autonomy Validation Platform & Pipelines at Zoox (link) via Christian Yang.

2026 Summer Intern, MS/PhD, ML Runtime and Deployment at Waymo (link) via Tracy F.

Staff ML Performance Engineer (Training Efficiency) at Wayve (link) via Paul Elvinger.

Here’s a full list of the jobs we’ve featured (link).

Job Moves

Chase Alexander: Software Engineer at Waymo (link).

Yasi Sedghani: Legal Director, Autonomous Mobility & Delivery Legal Team at Uber (link).

Cool Rides

‘My experience with a self driving car Waymo! 🚙’ (link).

‘Ride with Pony AI autonomous taxi. Very smooth ride - might be better than Waymo’ (link).

‘The ride in Waymo was not scary but exciting, the way the car manoeuvred the LA traffic and made it to the destination is an ode to human ingenuity, innovation and the never dying spirit of innovation and making lives better’ (link).

Shout-outs

Big thanks to TDD readers at Carvolution, and to Jane L and Rani 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.

We’re now over 8,000+ subscribers! Interested in advertising with TDD? Check out our media kit here and reach out to us.

Refer a friend

Until next week :)

-Harry

The Driverless Digest

Keep On Rolling: Electrifying, Scaling & Automating The Vehicles of Tomorrow (Curbivore 2026)

Chapters

Urban Autonomy in America: What’s Real & What’s Next (Curbivore 2026)

Chapters

Meet Dot: DoorDash’s Accelerated Autonomous Local Commerce Solution (Curbivore 2026)

Chapters

Tesla Launches Robotaxi Service in Dallas and Houston

Top Stories of the Week

Other Stuff

The Human Edge in AV: Scaling Swift Emergency Response

What else we’re reading/listening to

AV Events

AVs/Humans behaving badly

Neat Jobs

Job Moves

Shout-outs

Lucid Lands $750 Million in Funding, Including $200 Million from Uber

Top Stories of the Week

Other Stuff

Built for Autonomous Fleet Deployment

What else we’re reading/listening to

AVs/Humans behaving badly

AV/Humans behaving “goodly” :)

Neat Jobs

Job Moves

Shout-outs

Analyzing Waymo’s City Distribution Using Rider-Only Miles

Historical Rider-Only Miles Data

Total Rider-Only Miles

City Distribution Over Time

Quarter over Quarter Growth

Volkswagen's MOIA to Kick Off Robotaxi Testing in Los Angeles

Top Stories of the Week

Waymo InOut of NYC

Other Stuff

Faster Data Off the Vehicle—Faster AI on the Road

What else we’re reading/listening to

AVs/Human behaving badly

AV behaving “goodly” :)

Neat Jobs

Job Moves

Shout-outs

DoorDash’s Autonomous Delivery Strategy with Ashu Rege

Chapters

Designing Charging Hubs for Autonomous Fleets

Notes/Links:

China Robotaxi Outage Leaves Passengers Stranded on Highways

Top Stories of the Week

Other Stuff

The Human Edge in AV: Scaling Swift Emergency Response

What else we're reading/listening to

AVs behaving badly

AVs behaving “goodly” :)

Neat Jobs

Job Moves

Shout-outs

How Will London’s Driver Market Respond to Autonomous Vehicles?

What does London’s taxi market look like today?

The reasons why the next taxi conflict will not look like the last

The multiple disrupting entrants problem

The sovereign AI problem

When can we expect opposition?

Will the incumbents (Black cabs and ride hail drivers) coordinate between each other?

Who will the opposition target? What methods will they use? And what may they demand?

Conclusion

The Humans Powering Autonomous Vehicle Operations (Omar Zoubi, TaskUs)

Chapters

Notes/Links:

Waymo Hits 500,000 Weekly Rides and Over 4 Million Miles

Top Stories of the Week

Other Stuff

The Human Edge in AV: Scaling Swift Emergency Response

What else we’re reading/listening to

AVs/Humans behaving badly

AVs/Humans behaving “goodly” :)

Neat Jobs

Job Moves

Shout-outs

The AV Crisis Playbook