System Malfunction Freezes 100+ Robotaxis in Wuhan Traffic

A mass software malfunction in the central Chinese city of Wuhan caused more than 100 Baidu robotaxis to halt in the middle of busy roads this week, stranding passengers and disrupting traffic in one of the world’s most advanced testbeds for autonomous vehicles.

Image by Sky News

Robotaxis Stall Across Major Arteries in Wuhan

According to published coverage in Chinese and international media, the incident unfolded on the evening of March 31 in Wuhan, a city that has become a flagship hub for Baidu’s Apollo Go driverless taxi service. At around 8:57 p.m. local time, more than 100 autonomous vehicles reportedly froze in place on city streets and expressways after a centralized system malfunction triggered automatic safety stops.

Images and video circulating on Chinese social media, referenced in news reports, show the white Baidu-branded cars stopped in live lanes, some on multilane highways and elevated roads with traffic flowing around them. In several clips, hazard lights appear to be blinking as conventional vehicles carefully navigate past the immobilized robotaxis.

Local traffic bulletins and subsequent media reporting indicate that the failures were spread across multiple districts, affecting both downtown streets and outer-ring corridors. While the precise fault has not yet been detailed, publicly available information suggests that a software or network disruption propagated through the fleet almost simultaneously.

Reports indicate that Wuhan’s traffic police received a wave of calls from confused drivers and passengers as the incident unfolded. The large number of stalled vehicles and their locations in active lanes quickly turned what would normally be a routine technical glitch into a citywide traffic management challenge.

Passengers Stranded in Live Traffic but No Injuries Reported

Accounts summarized in outlets such as the Associated Press, Sky News, and other international media describe passengers being abruptly halted mid-journey, sometimes in the middle of fast-moving traffic. In at least one widely cited case, a passenger said their robotaxi stopped just after turning a corner, displaying an on-screen message that read “driving system malfunction” and indicating that staff were expected to arrive within minutes.

Some riders were able to open doors and exit safely, especially where cars had rolled to a stop along the curb or on slower streets. However, published reports note that others were reluctant to step out because their vehicles had stopped in center lanes with traffic passing on both sides, raising concerns about the risk of being struck by another car.

Publicly available information from local statements and media coverage indicates that, despite the scale of the disruption, no injuries were reported. Some outlets mention minor collisions and near misses as other drivers encountered the frozen vehicles unexpectedly, but there have been no reports of serious crashes linked directly to the malfunction.

The incident is being described in multiple reports as the first mass shutdown of driverless taxis of this magnitude in China. For many Wuhan residents, it appears to have been an unsettling reminder that while robotaxis are marketed as a safer, more predictable alternative to human drivers, large-scale technical failures can introduce new kinds of risk.

Baidu’s Apollo Go Under Spotlight in Its Largest Market

Baidu’s Apollo Go service is among China’s most prominent autonomous driving programs, operating commercial robotaxi services in several major cities. Wuhan is considered one of its most important markets, with publicly available company and government figures pointing to a fleet that has expanded to hundreds, and in some reports more than 1,000, driverless vehicles operating in defined zones.

The company has promoted Wuhan as a showcase for fully driverless operation, with many vehicles running without safety drivers and offering rides booked through a mobile app. Before this week’s malfunction, local coverage had highlighted rapidly growing ride volumes and extended operating hours as evidence that the technology was moving toward everyday integration.

Following the outage, Apollo Go’s technical architecture and safety redundancies are drawing renewed attention. Analysts cited in technology and business outlets note that the apparent simultaneity of the failures points to a shared dependency, such as cloud-based coordination, high-definition mapping services, or over-the-air software control, rather than isolated sensor problems in individual cars.

Publicly available reporting indicates that Baidu has not yet released a detailed technical explanation, and there is no confirmation to date of whether the issue was linked to an internal software bug, third-party infrastructure problem, or other external factor. What is clear from open-source information is that the company faces pressure to demonstrate that its safety protocols and fail-safe procedures are robust enough for large-scale deployment.

Safety, Regulation and Public Trust in Driverless Fleets

The Wuhan malfunction arrives at a sensitive moment for the global robotaxi industry. Similar high-profile disruptions have occurred elsewhere, including reports of self-driving vehicles in San Francisco coming to a halt during power outages or traffic light failures, prompting investigations and tightened oversight in some jurisdictions.

In China, authorities at national and municipal levels have actively encouraged autonomous driving trials, designating test zones and loosening rules in select districts to enable driverless operation. At the same time, recent policy documents referenced in industry analyses emphasize that commercial services must demonstrate clear safety benefits and maintain public confidence.

Observers quoted in technology and transport coverage suggest that the Wuhan incident may accelerate calls for clearer standards on how fleets are monitored and what must happen when central systems fail. Questions are likely to focus on whether cars should automatically move to the nearest safe stopping point, how long passengers can be left in stalled vehicles, and how easily occupants can exit and reach safety when traffic is still flowing around them.

There is also growing debate over whether highly centralized control architectures, which allow operators to update and coordinate thousands of vehicles at once, may create systemic vulnerabilities. If a single software bug, misconfiguration, or network outage can immobilize large portions of a city’s driverless fleet, regulators and the public may demand stronger safeguards and more independent fallback capabilities in each vehicle.

Implications for Global Expansion of Chinese Robotaxis

The malfunction in Wuhan is likely to resonate beyond China’s borders. Baidu has announced ambitions to export Apollo Go technology and services to overseas markets, including pilot projects in the United Kingdom and potential partnerships with international ride-hailing firms. Reports on those plans now sit alongside headlines about the Wuhan outage, complicating the narrative around rapid expansion.

Transport analysts and industry commentators note that regulators in Europe and North America are closely watching early deployments of Chinese autonomous vehicles. High-visibility failures can influence how cautious foreign agencies choose to be when granting test permits, setting operating conditions, or approving fully commercial services.

For travelers and residents in cities that are beginning to see robotaxis as part of their transport mix, the Wuhan episode underscores both the appeal and the uncertainty of the technology. Proponents point to potential reductions in crashes caused by human error, lower operating costs, and improved accessibility. Critics counter that rare but dramatic failures, especially those that unfold simultaneously across large fleets, may impose new types of systemic risk that are not yet fully understood.

In the short term, observers expect Baidu and local authorities to review response protocols, including how quickly staff can reach stalled vehicles and how passengers are guided to safety when unexpected stoppages occur. Longer term, the Wuhan malfunction is likely to become a case study in how cities manage the transition from experimental deployments to everyday reliance on driverless mobility.