Self-driving car: Has Europe already lost the race?

A race that has moved from prototypes to operations

The working headline is also the central question: Self-driving car: Has Europe already lost the race? The short answer is no, but the more uncomfortable answer is that Europe is now competing from behind. The evidence is no longer a glossy concept car or a staged demo; it is the appearance of commercial robotaxi services, fleet depots, regulatory fights and partnerships designed to keep vehicles earning revenue all day.

In the United States, the tempo is being set by three different models. Waymo is scaling by adding cities and relying on operational experience. Zoox is moving from testing toward a commercial service around a vehicle designed without the assumptions of a human driver. Tesla is attempting the most disruptive cost story with Cybercab, a purpose-built robotaxi that removes the steering wheel and pedals and leans on a camera-first autonomy strategy. Together, those moves show that the race has shifted from “can the car drive?” to “can the company deploy, clean, charge, insure, regulate and monetize thousands of cars?” [1]

That is precisely where Europe’s gap becomes visible. Europe has strong automakers, world-class suppliers and serious safety regulators. But the commercial scaling of autonomous ride-hailing is currently happening mainly in the United States and China, while Europe remains defined by pilots and regulatory preparation [1]. In other words, the continent is not absent from the race; it is running a different, slower event.

Tesla’s Cybercab changes the cost debate

Tesla’s Cybercab matters because it reframes autonomy as a manufacturing and operating-cost problem. A robotaxi with no driver only becomes economically transformative if the vehicle is cheap enough, available enough and easy enough to maintain. That explains why Tesla is not merely putting autonomous software into existing cars; it is trying to create a vehicle and a network around the economics of high utilization.

The Cybercab’s launch has already triggered a regulatory confrontation in the United States. Frandroid reported on September 8 that Tesla’s steering-wheel-free vehicle is under scrutiny from the National Highway Traffic Safety Administration, which wants to examine how Tesla judged that some federal safety standards were not applicable to a vehicle without human controls [2]. That matters for Europe because it illustrates the American advantage and risk at the same time: deployment can happen faster, but the legal argument may be tested after the vehicle is already on the road.

Tesla also appears to be preparing for a scale model that goes beyond a company-owned fleet. Axios reported that Tesla has begun soliciting interest from prospective partners for Cybercab fleet purchasing and for mobility hubs or infrastructure, while still leaving major economic details undisclosed, including the price of the Cybercab and the revenue split with operators [3]. That is not a full franchise system yet, but it is a signal: Tesla knows robotaxis are not only software. They are real estate, electricity, cleaning, repairs, insurance, finance and local operations.

A second Axios report described a proposed Fleetport concept: hubs that could store 100 to 250 robotaxis overnight while offering fast charging, cleaning and basic maintenance [4]. The detail may sound mundane, but that is the point. Once robotaxis move from demos to fleets, the bottleneck is no longer only artificial intelligence. It is the ability to make driverless cars behave like an industrial service.

Waymo and Zoox are building the operational benchmark

Tesla gets much of the attention, but Waymo currently defines the operational benchmark. ADT reported on September 8 that Waymo is carrying initial passengers in Denver, San Diego and Tampa and says its service is now available in 14 U.S. cities [1]. The same report said Waymo completes more than 500,000 paid rides per week in the United States and is regarded as the most advanced operator of commercial Level 4 robotaxis [1].

That lead is important because autonomy improves through exposure to edge cases, city-specific traffic patterns and passenger behavior. A company operating at scale learns about curb access, airport flows, poor weather, emergency vehicles, vandalism, rider support and fleet utilization. Those lessons are hard to simulate and harder to import after a rival has accumulated them.

Waymo’s strategy is also evolving through partnerships. Axios reported on September 9 that Lyft riders in Nashville can now be matched with Waymo robotaxis, while Lyft’s Flexdrive unit will handle charging, cleaning and maintenance for hundreds of Waymo vehicles at a new 80,000-square-foot depot [5]. This shows the maturity of the market: the winning company may not be the one with the cleverest demo, but the one with the best operating web of apps, depots, capital and city relationships.

Zoox represents a different bet. According to ADT, the Amazon-backed company is expanding testing in Houston and San Diego and preparing its own driverless robotaxi, designed without a steering wheel or pedals [1]. ADT also noted that Zoox became the first robotaxi company to receive a commercial exemption from the U.S. safety regulator for its purpose-built vehicle, allowing it to charge fares and move from pilot operation toward commercial service [1]. That distinction is crucial: the future robotaxi may not be a modified car at all, but a small urban transport appliance.

Europe’s problem is not talent; it is timing

Europe has not “lost” because it lacks competence. Its manufacturers understand safety, industrialization, homologation and large-scale production. Its suppliers have deep expertise in sensors, braking, steering, chips, interiors and fleet systems. Its cities are dense enough to make shared autonomous mobility useful. The weakness is the translation of capability into live commercial services.

The regulatory logic is different on each side of the Atlantic. Frandroid emphasized that in Europe, vehicle approval generally happens before deployment, while the U.S. system allows manufacturers to self-certify and then face scrutiny later [2]. That upstream European caution protects the public from premature rollouts, but it also slows the learning loop. If the United States learns by deploying and Europe deploys only after lengthy harmonization, the performance gap can widen even when European rules are more coherent.

France illustrates the dilemma. A cautious approach can be rational: autonomous vehicles must prove safety, cybersecurity, accountability and insurance clarity before mass adoption. But if caution becomes industrial passivity, the country risks discovering that the key platforms, operating systems, sensor stacks and fleet interfaces have already been standardized elsewhere.

The situation is not hopeless. ADT reported that Waymo plans to enter Germany, with mapping and test drives in Munich expected to begin in the coming weeks and a public robotaxi service planned for late 2027 [1]. That means Europe can still become a major deployment geography. The open question is whether European companies will own the platform or host it.

China adds another layer of pressure

The comparison is not only Europe versus America. Chinese autonomous-driving companies are also pushing beyond their home market. Pony.ai announced on September 9 that it operates a fare-charging commercial robotaxi service in Doha with Mowasalat and that Doha is part of its international expansion strategy alongside China [6]. The same announcement said Pony.ai has operations spanning China, Europe, East Asia and the Middle East, and that its planned and potential overseas deployments exceed 4,000 vehicles [6].

For Europe, that creates a two-front challenge. U.S. firms lead in high-profile commercial robotaxi branding and deployment. Chinese firms bring aggressive commercialization, manufacturing scale and partnerships outside China. European automakers may still be excellent at building cars, but robotaxis are becoming platforms, not products.

Has Europe already lost?

Europe has not lost the self-driving car race in the way a team loses a final match. There is no single finish line. But it has already lost the first phase of commercial momentum. The United States has companies proving operations at scale. China has companies pushing mass deployment and exporting autonomy. Europe still has safety credibility, industrial depth and urban demand, but it lacks a clear champion operating robotaxis at comparable scale.

The real danger is not that Europeans will never ride in autonomous vehicles. They will. The danger is that the vehicles, software, data models and operating platforms will be imported, while Europe’s role is reduced to approval, assembly and local adaptation. If that happens, Europe will not have lost the car. It will have lost the operating system of urban mobility.

To recover, Europe needs faster test corridors, clearer Level 4 approval paths, public procurement for autonomous shuttles and robotaxis, and a willingness to let cities experiment under strict safety reporting. It also needs its automakers to stop treating autonomy as an optional feature package and start treating it as a new transport industry.

For now, the race is still open. But the leaders are no longer waiting at the starting line.

Sources from the last 72 hours

1. [1]Waymo and Zoox expand US robotaxi raceSep 8, 2026, 3:31 PM UTC
2. [2]Cybercab sans volant : Tesla utilise la stratégie « foncer d'abord, réguler ensuite », la NHTSA enquêteSep 8, 2026, 10:10 AM UTC
3. [3]Entrepreneurs could hold the key to Tesla's Cybercab plansSep 9, 2026, 4:00 PM UTC
4. [4]Cybercab fleets would need a place to sleepSep 9, 2026, 4:00 PM UTC
5. [5]Why Lyft could be Waymo's ideal ride-hailing partnerSep 9, 2026, 10:00 AM UTC
6. [6]PONY AI Inc. Highlights Progress of Doha Robotaxi Commercial Operations at AEMOB ForumSep 9, 2026, 4:30 AM UTC

AI-generated article based on recent web research, then preserved as a dated editorial snapshot.