Waymo Now Runs Paid Robotaxis in 14 Cities and Drivers Are Doing the Math

Waymo began offering paid robotaxi rides to the public in Denver, San Diego and Tampa on Tuesday, taking the Alphabet-owned company's commercial operations to 14 US cities, while Amazon-owned Zoox said it will start testing in Houston and San Diego, Quartz reported. At Metaintro, we track this because roughly 455,700 people in the US drive for a living in the taxi, shuttle and ride-hailing category, and the question that matters to them is not whether robotaxis work. It is how fast the map fills in, and the answer so far is city by city rather than everywhere at once.

What exactly did Waymo launch?

Paid public rides in three new cities on Tuesday, hailed through the Waymo app, with the company saying it will open access gradually in each city as it has done in previous markets. Tens of thousands of people had already signed up in each city before launch. Each market starts with dozens of vehicles and the company plans to reach hundreds over time.

The scale behind that is more informative than the city count. Waymo's fleet now totals more than 4,000 vehicles across its US markets according to TechCrunch, and the company said it currently completes more than 500,000 paid trips per week with a goal of 1 million weekly rides before the end of the year. Denver and San Diego will be served only by the Ojai, a Zeekr-built minivan running Waymo's sixth-generation self-driving system, while Tampa will also have Jaguar I-Pace vehicles. Denver matters technically because it is a cold-climate market and the Zeekr-built Ojai is engineered for winter road conditions.

That detail is the one to watch if you want to know how far this goes. Autonomous services have concentrated in warm, dry, grid-planned cities because those conditions are the easiest to solve, and snow, ice and poor lane visibility have been the practical ceiling on expansion rather than any regulatory barrier. A cold-weather market operating through a full winter would remove the main technical argument for why large parts of the country were going to be slow to see this. A cold-weather market that struggles would confirm it.

Where does Zoox actually stand?

Considerably earlier in the process, and the difference is worth understanding because coverage often treats the two as equivalent. Zoox's Houston and San Diego testing will include safety drivers in the vehicles, which means humans are still employed in the loop. According to CNBC, Las Vegas remains the only market where Zoox charges passengers, and in San Francisco it offers a complimentary service to an invited group of users. The expansion would take its presence to 12 US markets, but presence and paid commercial service are not the same thing.

That distinction is the one drivers should watch. A market with safety drivers is a market that has added jobs rather than removed them, temporarily. Federal oversight sits with the National Highway Traffic Safety Administration and the wider Department of Transportation, and permitting varies enough by state and city that the same company can be years further ahead in one metro than in another two hundred miles away. The transition point that affects driver employment is not when testing begins, it is when the safety driver comes out and paid rides open to the public, and those two events can be years apart in a given city.

That gap is the single most useful thing for a working driver to understand, because most alarming coverage is triggered by the first event and most economic effect follows the second. A city can carry headlines about autonomous vehicles arriving for two or three years while every one of those vehicles still has a paid human sitting in it. Reading the announcements without that filter produces a sense of imminent replacement that the operational reality does not support, and panic is an expensive basis for a career decision.

Does this actually mean fewer driving jobs?

Not yet, and the official projections say something surprising. The Bureau of Labor Statistics counts 455,700 taxi drivers, shuttle drivers and chauffeurs in 2025 at a median wage of $37,960 a year, and projects employment growing 9 percent between 2025 and 2035, described as much faster than the average for all occupations, adding about 41,400 jobs.

Both things can be true at once because the effects are geographically concentrated. Robotaxi displacement is intense inside a handful of dense urban cores and absent everywhere else, while demand for paid rides keeps growing across the whole country. A driver in San Francisco or Phoenix is experiencing something a driver in a mid-sized city is not, and the national aggregate averages the two into a number that describes neither. The same pattern shows up internationally, which is why China has been pumping the brakes on robotaxis specifically to protect driver jobs.

There is a second reason the national projection stays positive. Ride-hailing demand has been growing for a decade, and automated capacity is being added into a market that is expanding rather than a fixed one, so the first effect is usually more rides served rather than fewer drivers working. Displacement arrives when automated supply growth outpaces demand growth in a given city, which is a slower and more local process than a launch announcement suggests.

How quickly is the map actually filling in?

Faster than the city count suggests, and slower than the coverage implies. Fourteen cities in commercial operation with more than 4,000 vehicles is a real business rather than a pilot, and a target of doubling weekly rides to 1 million inside a year is an aggressive ramp. But dozens of vehicles per new city, growing to hundreds, is a small share of the ride volume in any large metro, and the gradual access rollout means even a launched city is not fully served for some time.

The more useful metric for a driver is trips per week in your specific city rather than the national fleet size. Waymo entering a market does not immediately change your earnings, but sustained growth in its local trip share does, and that shows up first as longer waits between fares at the times and places the robotaxis serve best. Late evening weekend trips in a compact downtown are the clearest early indicator, because that is both the most profitable slot for a human driver and the easiest one for an automated service to cover. Anyone who watched the previous phase of this in Dallas has seen the sequence before, covered in Waymo opening its Dallas service to everyone.

What does this mean for your career?

If you drive for a living, the practical question is which segment you are in rather than whether automation is coming. Short urban trips in dense, well-mapped, temperate city centres are the most exposed, because that is exactly the operating envelope these services are built for. Airport runs, long-distance and intercity work, non-standard vehicles, accessibility transport, corporate and executive driving, and anything requiring loading, assistance or a licence beyond a standard one are considerably less exposed for now. The test worth applying is to your own week rather than to the occupation as a whole. If the value you add on a given trip is entirely moving the vehicle from one point to another, that trip is exposed. If it includes lifting, securing a wheelchair, managing an itinerary, reassuring a nervous passenger or knowing a site's access rules, it is not, and those trips are worth deliberately increasing as a share of your work now rather than after the local market shifts.

The move that follows from that is lateral rather than dramatic. Commercial licences open into segments where the vehicle carries goods or many passengers rather than one, and those show different projections and different pay, which you can check directly across the transportation and material moving occupations pages. "In combination with layoffs over the recent years plus the rise of required AI skills, experience is no longer enough," Lacey Kaelani, chief executive of Metaintro, told Fortune. For driving work the equivalent is that a standard licence is no longer enough on its own, while an endorsement, a speciality or a segment that needs a person present is.

Are new jobs being created alongside this?

Yes, though fewer and different. A robotaxi fleet needs depot operations, cleaning, charging, maintenance, remote assistance, fleet response and mapping work, and those roles are real and locally based. They are also fewer per vehicle than a driver-operated fleet by design, since removing labour cost is the commercial rationale, so this is not a like-for-like substitution in headcount terms.

The honest way to describe the trade is that it converts a large number of flexible, low-barrier, self-scheduled jobs into a smaller number of fixed-shift technical ones. That matters because the flexibility was the point for many drivers, who chose the work precisely because it fitted around study, caring responsibilities or another job. A depot technician role is a better job on most measures and a worse fit for that specific need, which is a real loss that a headline employment number will not capture. It is also a job with a different entry requirement. Driving needs a licence and a short induction, while fleet maintenance and remote assistance need technical training, which means the transition is not automatic for the people displaced even where the roles exist in the same city. Employers running these fleets could close that gap deliberately, and the ones that do will have a considerable advantage in local goodwill when they arrive in a new market. The wider gig economy has been absorbing similar shifts, including when Amazon shut a long-running gig work platform.

What should drivers do in the next year?

Watch three things rather than the news cycle. First, whether a service has launched paid public rides in your city as opposed to testing, since only the former competes with you. Second, whether safety drivers have been removed, which is the real commercial threshold. Third, your own earnings by time of day and pickup zone over a few months, because local displacement shows up in your own data long before it appears in any statistic.

There is a fourth thing worth tracking, which is fare pressure rather than trip volume. An operator with no labour cost can sustain lower prices than a human driver can, and the first competitive effect in several markets has been on what a ride is worth rather than on how many rides exist. If your earnings per trip are falling while your trip count holds, that is the same phenomenon arriving through a different door, and it is easy to misread as a platform pricing change.

Then use the time. Driver employment is projected to grow rather than shrink nationally through 2035, which means most drivers have years rather than months, and that is enough time to add a commercial endorsement, move into a less exposed segment or retrain entirely without panic. Anyone whose platform relationship is already precarious should also understand their rights, since gig platforms cutting people off with no appeal remains a more immediate risk to most drivers' income this year than any robotaxi.

Related Articles

• Transportation Jobs Slow Down While Warehousing Sees Growth
• New Jersey Sues Amazon Over Delivery Driver Wages and Route Control
• Surge Pricing Is Quietly Squeezing Gig Workers, Here Is the Fix
• The Labor Department Wants Gig Workers Buying Health Insurance Through Associations
• Nine High-Paying Transportation Careers
• Drone Delivery Startup Manna to Add Jobs at a New Tulsa Facility
• Trucking Jobs Blocked as Kroger and Carriers Face Antitrust Claims
• Walmart Now Delivers Subway in Minutes, Inside the Gig Work Push
• How to Become a Transportation Manager
• Job Losses Loom as AI-Powered Robots Gain Traction in Agriculture and Construction

People Also Asked

Q: How many cities does Waymo operate in?

A: Fourteen US cities for paid commercial service after launching in Denver, San Diego and Tampa. The fleet totals more than 4,000 vehicles across those markets, each new city starts with dozens of vehicles and grows to hundreds, and the company reports more than 500,000 paid trips per week with a target of 1 million weekly rides before the end of the year.

Q: Are robotaxis reducing the number of driving jobs?

A: Not at a national level yet. The Bureau of Labor Statistics counts 455,700 taxi, shuttle and ride-hailing drivers and chauffeurs in 2025 and projects 9 percent employment growth through 2035, adding about 41,400 jobs. Displacement is concentrated in a small number of dense urban markets rather than spread across the country.

Q: Which driving work is least exposed to automation?

A: Work that falls outside the dense, well-mapped, short-trip urban envelope these services target. That includes long-distance and intercity driving, accessibility and assisted transport, corporate and executive driving, and any role requiring loading, passenger assistance or a commercial licence. Segments where a person must be physically present for reasons beyond steering are the most durable. That is the test worth applying to your own week rather than to the occupation as a whole. If the value you add on a given trip is entirely getting the vehicle from one point to another, that trip is exposed. If it includes lifting, securing a wheelchair, managing an itinerary, handling a nervous passenger or knowing a site's access rules, it is not, and those trips are worth deliberately increasing as a share of your work now rather than after the local market changes.

Driving work is not disappearing nationally, but it is changing shape city by city, and the drivers who move segment early are the ones who choose their next job rather than take what is left. Metaintro indexes live transportation and logistics roles across every licence class so you can see what your experience already qualifies you for. Create your free profile and find the routes still hiring people.