Waymo provides 500,000 weekly paid robotaxi trips across 10 U.S. cities as of March 2026 (per TechCrunch ridership analysis), while Tesla operates only 25 unsupervised Model Y vehicles across three Texas cities in the same window, according to NHTSA Standing General Order filings. The gap between the two American leaders shows that scale, not technology demos, defines the current robotaxi market. Apollo Go has provided 20 million rides worldwide cumulatively by February 2026, according to Baidu corporate communications, pushing the company past every Western competitor in raw trip volume.
Key Takeaways
- Waymo’s average weekly paid robotaxi trips have grown tenfold, from 50,000 in May 2024 to 500,000 in March 2026.
- Waymo achieved a 90% reduction in serious injury crashes across 127 million miles of autonomous operation.
- Apollo Go fleets accumulated 190 million kilometers in fully driverless mode, with airbags deploying once per 12 million kilometers on average.
- Pony.ai’s Q4 2025 robotaxi revenues rose 159.5% year-over-year to $6.7 million, with fare-charging revenue surging over 500%.
- Tesla’s median robotaxi ride wait time was 15.32 minutes, compared with 5.74 minutes for Waymo and 3.15 minutes for Uber.
- Six carriers hold California Public Utilities Commission autonomous vehicle permits, while Tesla holds only a Transportation Charter Party limousine permit.
- GM had invested $10 billion in Cruise before halting its robotaxi program in December 2024.
Editor’s Choice
- Waymo’s $16 billion investment round valued the company at $126 billion post-money in February 2026.
- In 2025, Waymo’s annual trips totaled 14 million, approximately three times the 4.5 million rides provided in 2024.
- Pony.ai’s robotaxi fleet exceeded 1,400 units as of March 25, 2026, with a target of 3,000+ vehicles by year-end.
- Apollo Go delivered 3.4 million robotaxi trips in the 4th quarter of last year, a 200% year-over-year increase.
- Zoox has surpassed nearly two million autonomous miles with over 500,000 people on its waitlist.
- By July 2025, Waymo reached 100 million fully autonomous miles, with weekly autonomous mileage exceeding 2 million miles by year-end.
- Apollo Go now operates across 26 cities worldwide as of February 2026.
Recent Developments
- March 26, 2026: Waymo crossed 500,000 weekly paid robotaxi trips across 10 U.S. cities.
- February 2, 2026: Waymo raised a $16 billion investment round at a $126 billion post-money valuation led by Dragoneer, DST Global, and Sequoia Capital.
- April 18, 2026: Tesla launched a fully unsupervised Robotaxi service in Dallas and Houston.
- March 31, 2026: Over a hundred Apollo Go robotaxis simultaneously froze on the streets of Wuhan, prompting Beijing to suspend new autonomous driving permits nationwide.
- March 26, 2026: Pony.ai reported Q4 2025 robotaxi revenue of $6.7 million, up 159.5% year-over-year.
- March 24, 2026: Zoox more than doubled the number of ride locations in Las Vegas and quadrupled its San Francisco service area.
Global Robotaxi Market and Industry Scale
The active robotaxi industry spans seven commercial operators across three continents. Waymo achieved a Series C valuation of $45 billion post-money in October 2024, then raised a $16 billion investment round valuing the company at $126 billion post-money in February 2026, a near-tripling of valuation in roughly 16 months.
| Operator | Parent | Service Type | Active Cities (2026) | Fleet Estimate |
|---|---|---|---|---|
| Waymo | Alphabet | Driverless commercial | 10 U.S. cities | ~3,000 vehicles |
| Tesla Robotaxi | Tesla | Unsupervised pilot | 3 Texas cities | 25 vehicles |
| Apollo Go | Baidu | Driverless commercial | 26 cities globally | 1,000+ vehicles |
| Pony.ai | Pony AI Inc. | Driverless commercial | 8+ cities | 1,400+ vehicles |
| WeRide | WeRide Inc. | Mixed driverless + drivered | UAE + China + US | ~750 robotaxis |
| Zoox | Amazon | Driverless commercial | 2 cities + 2 launching | <100 vehicles |
| Cruise | GM (defunct) | Halted Dec 2024 | None | None |
Source: Waymo investor disclosures, Pony AI Q4 2025 earnings, Baidu corporate communications, NHTSA Standing General Order filings, CPUC permit registry
Waymo Trips, Cities, and Fleet Statistics
Waymo dominates Western robotaxi volume by a margin most observers underestimate. In May 2024, Waymo recorded 50,000 weekly paid robotaxi trips, and by March 2026, that figure had grown to 500,000 weekly paid robotaxi trips. In 2025, Waymo’s annual trips totaled 14 million, approximately three times the 4.5 million rides provided in 2024. Waymo plans ride-hailing operations in over 20 additional cities in 2026, including Tokyo and London.
By the numbers: Waymo provides 500,000 weekly trips across 10 U.S. cities. Across 127 million miles of autonomous operation, Waymo achieved a 90% reduction in serious injury crashes. Waymo riders helped avoid over 18 million kilograms of CO2 emissions in 2025, with over 3.8 million rider hours logged. The fleet grew from approximately 1,500 vehicles in April to about 2,500 by November of last year.
- During 2025, Waymo’s weekly rides increased from 175,000 to over 450,000 weekly paid rides.
- Riders accumulated 3.8 million hours in Waymo vehicles throughout the year.
- By July 2025, Waymo reached 100 million fully autonomous miles in cumulative driverless operation, with weekly mileage exceeding 2 million miles by year-end.
- Waymo riders helped avoid over 18 million kilograms of CO2 emissions, a 3x increase from the prior year.
- Waymo maintains approximately 3,000 robotaxis equipped with its 5th-generation autonomous driving system.
- Waymo provides more than 400,000 rides every week across six major U.S. metropolitan areas as of February 2026.
| Waymo Metric | Value | Source Period |
|---|---|---|
| Weekly paid trips | 500,000 | March 2026 |
| Annual trips 2025 | 14 million | Full year 2025 |
| Annual trips 2024 | 4.5 million | Full year 2024 |
| Lifetime trips | 20 million+ | By December 2025 |
| Active US cities | 10 | March 2026 |
| Fleet vehicles | ~3,000 | March 2026 |
| Cumulative autonomous miles | 127 million | February 2026 |
Source: Waymo blog, Waymo investor disclosures, TechCrunch ridership analysis
Tesla Robotaxi Deployment and Fleet Size
Tesla’s commercial Robotaxi service occupies a different end of the same chart: small fleet, aggressive pricing, faster vehicle-replacement cycle. The Tesla Robotaxi service launched in a limited capacity in Austin, Texas, on June 22, 2025. Tesla operates 25 unsupervised Model Y vehicles across three Texas cities as of late April 2026, with 19 in Austin, 3 in Dallas, and 3 in Houston. As of December 2025, Tesla operated approximately 135 robotaxis in service, with about 29 in Austin and 106 in the San Francisco Bay Area for employee testing, a meaningful gap between supervised fleet count and unsupervised deployment.
Tesla aims to launch unsupervised robotaxi services in five additional cities. Phoenix, Miami, Orlando, Tampa, and Las Vegas, by mid-2026. The purpose-built Cybercab is expected to begin volume production in April 2026.
Apollo Go Operations in China and International Markets
Baidu’s Apollo Go runs the largest commercial driverless ride-hailing service by cumulative trips. Apollo Go has provided 20 million rides worldwide cumulatively by February 2026. The fleet has accumulated 300 million kilometers (186 million miles) of autonomous driving in total, with 190 million kilometers (118 million miles) in fully driverless mode. In its peak week, the company provided 300,000 such trips.
Key finding: Baidu Apollo Go: 3.4 million robotaxi trips in Q4 2025 alone, a 200% year-over-year increase, with airbags deploying once per 12 million kilometers on average. The platform now operates across 26 cities worldwide.
- Apollo Go operates with full commercial service in Beijing, Chongqing, Wuhan, and Shenzhen.
- On November 29, 2024, the Hong Kong Transport Department granted Apollo Go its first autonomous vehicle pilot license.
- On August 4, 2025, Baidu announced a partnership with Lyft to provide self-driving cars in Europe starting in 2026 in Germany and Britain.
- Recent international launches include Dubai, Abu Dhabi, and a UK London pilot planned for H1 2026.
- On March 31, 2026, over a hundred Apollo Go robotaxis simultaneously froze on the streets of Wuhan, stalling on overpasses and elevated roads.
- Beijing suspended all new autonomous driving permits nationwide following the Wuhan paralysis incident.
Pony.ai Revenue Growth and Unit Economics
Pony.ai is the first to report city-wide unit economics breakeven. In Q4 2025, Pony.ai robotaxi revenues were $6.7 million, representing a 159.5% increase from $2.6 million in the prior-year quarter. Fare-charging revenues surged over 500% year-over-year. Full-year robotaxi revenues totaled $16.6 million, up 128.6% from 2024.
Pony.ai attained consecutive break-even unit economics in Guangzhou and Shenzhen within four months of Gen-7 launch. Peak daily net revenue per Gen-7 vehicle reached an all-time high of RMB394, with 25 orders per vehicle.
Why it matters: Pony.ai’s city-wide breakeven within four months of Gen-7 launch is the first robotaxi profitability signal without subsidy.
- Pony AI’s robotaxi fleet size exceeded 1,400 units as of March 25, 2026, with a target to exceed 3,000 by year-end 2026.
- Pony.ai operations include Shenzhen, Guangzhou, Hangzhou, Changsha, Croatia, Doha, Dubai, and Singapore.
- The goal is to deploy Robotaxis in more than 20 cities globally by year-end 2026.
- Q4 2025 robotaxi revenue of $6.7 million reflected a 159.5% year-over-year jump from $2.6 million in Q4 2024.
WeRide Cross-Border Deployment Statistics
WeRide built its market position on cross-border licensing rather than home-market scale. WeRide holds a CPUC Drivered Pilot and Driverless Pilot permit in California, placing it in the small group of operators cleared by the regulator alongside Waymo and Zoox. WeRide also holds a landmark Abu Dhabi commercial license, described in industry coverage as the world’s first urban Level 4 commercial license granted outside the United States, and operates with safety drivers through Tawasul before transitioning to fully driverless service in early 2026.
Reported industry coverage places WeRide’s global autonomous fleet above 1,600 vehicles in Q3 2025, including roughly 750 robotaxis. The split fleet model (robotaxis plus robobuses and robosweepers) gives the company hardware diversification.
Zoox Las Vegas and San Francisco Service Footprint
Amazon’s Zoox built its early footprint on a single-vehicle design and a pair of pilot cities. Zoox has more than doubled the number of ride locations in Las Vegas, now including the Convention Center, major hotels, and the Strip. The San Francisco service area is being quadrupled for Explorers Program members, extending across the eastern half of the city, including the Marina, North Beach, Chinatown, Pacific Heights, and the Embarcadero. Austin, Texas, and Miami, Florida, are the two new cities receiving Zoox’s purpose-built robotaxis.
The company has surpassed nearly two million autonomous miles traveled and has over 500,000 people on its waitlist.
| Zoox Metric | Value | Period |
|---|---|---|
| Cumulative autonomous miles | ~2 million | March 2026 |
| Public waitlist | 500,000+ | March 2026 |
| Active service cities | 2 (Vegas + SF) | March 2026 |
| Announced new cities | Austin, Miami | 2026 |
| Las Vegas service launch | September 2025 | Free pilot |
Source: Electrek expansion coverage citing Zoox journal entries
Cruise Shutdown and GM’s Strategic Pivot
Cruise’s collapse in late 2024 reset capital-intensity expectations. GM announced on December 11, 2024, that it would stop development on its Cruise robotaxis. CEO Mary Barra stated the company had already invested $10 billion in Cruise and noted that launching and operating a robotaxi business was expected to require a significant amount of incremental time and capital beyond that figure. Barra emphasized that a robotaxi business is not General Motors’ core business, citing high capital costs as a factor.
The most serious incident occurred on October 2, 2023, when a Cruise vehicle struck a pedestrian who became trapped underneath. The vehicle then dragged the injured person approximately 20 feet at low speed. Following this incident, California’s Department of Motor Vehicles suspended Cruise’s operating permit, and the company also halted operations in Phoenix, Houston, and Austin.
GM cut Cruise spending by approximately $1 billion in January 2024, a 50% reduction. Sterling Anderson, formerly Tesla’s Autopilot chief, joined GM and outlined the restart plans during an August 6 employee meeting, pivoting the company toward personal-AV development rather than commercial robotaxi service.
Key finding: Cruise: $10 billion invested over the program’s lifetime, with the October 2, 2023 pedestrian-dragging incident in San Francisco marking the regulatory turning point. The company halted operations across three additional cities. Phoenix, Houston, and Austin, within weeks.
Robotaxi Safety Statistics from NHTSA Data
Safety data through 2026 splits the operator field into two camps: those publishing detailed crash narratives and Tesla. Across 127 million miles of autonomous operation, Waymo achieved a 90% reduction in serious injury crashes. Waymo achieved more than a 10-fold reduction in serious injury or worse crashes compared to human drivers across its 2025 operating window.
Tesla’s safety profile reads differently. Tesla’s robotaxis have been involved in 14 crashes in Austin, Texas, since the service launched in the city last summer. The five most recent crashes occurred during December and January. The crashes caused property damage when robotaxis collided with other vehicles or fixed objects, and no serious injuries were reported across the 14 incidents. Early in the rollout, NHTSA investigated reports of robotaxis driving erratically, including instances of driving on the wrong side of roads and sudden braking.
The transparency gap matters. Waymo, Zoox, Aurora, and Nuro publish detailed crash narratives in NHTSA filings; Tesla redacts its narratives, limiting third-party safety analysis.
Apollo Go safety performance shows airbags deploy once per 12 million kilometers on average, offering a different, and arguably more useful, denominator than per-mile crash counts.
| Operator | Safety Metric | Value |
|---|---|---|
| Waymo | Serious injury reduction vs human | 90% (over 127 million miles) |
| Waymo | Crash reduction vs human (serious or worse) | 10x+ |
| Tesla | Crashes in Austin since June 2025 | 14 |
| Tesla | Recent crashes (Dec 2025 to Jan 2026) | 5 |
| Apollo Go | Airbag deployment frequency | Once per 12 million km |
Source: Waymo investor disclosures, NHTSA Standing General Order filings, Baidu corporate communications
Robotaxi Fare Pricing vs Uber and Lyft
Rideshare-pricing analyst Obi published the cleanest cross-platform pricing comparison of the cycle. Average prices in the Bay Area between November 27, 2025, and January 1, 2026: Waymo at $19.69, Uber at $17.47, and Lyft at $15.47. Compared to June 2025, Waymo’s cost has dropped 3.62%, while Uber’s went up 12%, and Lyft’s climbed 7%. Obi conducted the analysis by simulating more than 94,000 ride requests in the Bay Area.
The takeaway: Waymo: Bay Area average ride cost has dropped 3.62% since June 2025 while Uber prices climbed 12% and Lyft prices climbed 7%. The price-trend lines are now narrowing toward each other rather than diverging, a structural shift that changes the consumer-substitution math.
The wait-time data tells the trade-off story more clearly. Average wait times: Tesla 15.32 minutes, Waymo 5.74 minutes, Lyft 5.14 minutes, and Uber 3.15 minutes. Tesla’s pricing edge is offset by a wait window roughly 5x that of Uber, suggesting the cheaper-than-Uber positioning depends on riders willing to swap minutes for dollars.
CPUC Permits and US Regulatory Framework
The California Public Utilities Commission permit registry is the authoritative public ledger of which operators can run a paying robotaxi service in the state. Current carriers reporting to the CPUC include Aurora Operations Inc., Nuro Inc., Tensor Auto Inc., Waymo LLC, WeRide Corp., and Zoox Inc. Waymo holds four permits. Drivered Deployment, Driverless Pilot, Drivered Pilot, and Driverless Deployment are the most comprehensive coverage of any operator. Tesla holds a Transportation Charter Party (TCP) permit, the same authorization as a limousine company in California, which is not an autonomous vehicle permit.
- Only Driverless Deployment permittees are required to provide ODD information under the CPUC framework.
- Waymo Driverless Deployment includes Bay Area and Los Angeles Area maps with a separate Operational Design Domain Statement document.
- WeRide Corp. holds Drivered Pilot and Driverless Pilot permits in California.
- Zoox Inc. holds Drivered Pilot and Driverless Pilot permits in California.
- Aurora, Nuro, and Tensor Auto each hold a single Drivered Pilot permit in the state.
EU and UNECE Autonomous Vehicle Regulation
The international regulatory picture moved decisively in early 2026. The UN Economic Commission for Europe’s Working Party on Automated/Autonomous and Connected Vehicles (GRVA) approved a Global Technical Regulation on Automated Driving Systems in late January 2026. This regulatory framework took approximately a decade to finalize. The next WP.29 meeting session, at which the AC.3 approval could take place, is scheduled for June 23 to 26, 2026.
The GTR employs a safety assurance approach rather than prescriptive performance metrics, replacing prescriptive testing requirements with structured documentation. Key structural elements include a Safety Management System, a Safety Case, an Operational Design Domain, In-Service Monitoring and Reporting requirements, and a Data Storage System for ADS recording safety-related performance data.
For European deployment, Waymo and Baidu both target 2026 launches in Western European cities, with national regulators in the UK and Germany having completed enabling legislation.
Passenger Demographics and Adoption Willingness
Survey data on robotaxi adoption shows wider variance by demographic than by city: men report roughly twice the willingness rate of women, riders under 30 skew strongly positive, and college-grad cohorts outpace high school cohorts.
Ecosystem familiarity predicts adoption: the cohort that grew up summoning Uber is the least surprised by a steering wheel without hands on it.
Parent Company Investment Levels and Funding
The robotaxi capital stack now reads like an infrastructure-fund roster. Waymo raised a $16 billion investment round valuing the company at $126 billion post-money. The financing was led by Dragoneer Investment Group, DST Global, and Sequoia Capital, with significant investments from Andreessen Horowitz and Mubadala Capital. T. Rowe Price, Bessemer Venture Partners, Silver Lake, Tiger Global, BDT and MSD Partners, CapitalG, Fidelity Management and Research Company, GV, Kleiner Perkins, Perry Creek Capital, and Temasek also participated. Alphabet maintained its role as the majority investor.
Waymo’s ridership remains a sliver of Uber’s human-driven ride-hailing business, which completed over 1 million mobility trips per hour as of August 2024, a useful reality check on absolute scale despite the funding momentum.
Frequently Asked Questions (FAQs)
Waymo provides 500,000 weekly paid robotaxi trips across 10 U.S. cities as of March 2026. The number grew tenfold from 50,000 weekly trips in May 2024, representing one of the fastest scale curves in commercial transportation. By the end of 2026, the company is on track to reach this volume weekly across its current and announced markets.
Tesla operates 25 unsupervised Model Y vehicles across three Texas cities as of late April 2026, with 19 in Austin, 3 in Dallas, and 3 in Houston. Tesla also runs a larger fleet of 106 vehicles in the San Francisco Bay Area for employee testing, with those vehicles operating under safety-monitor supervision.
Chinese operators hold the volume lead by cumulative trips. Apollo Go alone reported 20 million worldwide rides by February 2026, while Pony.ai operates a fleet exceeding 1,400 units. 500,000 weekly paid robotaxi trips spread across approximately 3,000 vehicles anchor Waymo’s US per-vehicle utilization lead.
Waymo achieved a 90% reduction in serious injury crashes across 127 million miles of autonomous operation. Waymo achieved more than a 10-fold reduction in serious injury or worse crashes compared to human drivers. Tesla’s robotaxis have been involved in 14 crashes in Austin, Texas, since the service launched, with no serious injuries reported.
GM announced on December 11, 2024, that it would stop development on its Cruise robotaxis after investing $10 billion. Barra emphasized that a robotaxi business is not General Motors’ core business, citing high capital costs. An October 2, 2023, pedestrian-dragging incident in which the vehicle dragged the injured person approximately 20 feet at low speed triggered the California DMV permit suspension.
Carriers reporting to the CPUC include Aurora, Nuro, Tensor Auto, Waymo, WeRide, and Zoox. Waymo holds four permits (Drivered Deployment, Driverless Pilot, Drivered Pilot, and Driverless Deployment), the most comprehensive coverage of any operator. Tesla holds a Transportation Charter Party (TCP) permit, the same authorization as a limousine company in California, which is not an autonomous vehicle permit.
Conclusion
Robotaxi statistics through early 2026 show a market splitting into three tiers: scaled commercial operators, pilot-stage operators, and exited operators. Waymo provides 500,000 weekly paid robotaxi trips across 10 U.S. cities and has logged 127 million miles of autonomous operation. Apollo Go has provided 20 million rides worldwide cumulatively by February 2026, and Pony.ai’s city-wide unit economics mark the structural maturation of Chinese operators. Tesla operates 25 unsupervised Model Y vehicles across three Texas cities, a scale outlier among US peers despite headline coverage volume.
Waymo raised a $16 billion investment round valuing the company at $126 billion post-money, putting the company in the same range as legacy automakers, while GM had invested $10 billion in Cruise before exiting. For riders, regulators, and adjacent industries, the data points toward a market where the next two years determine which operators reach durable scale.