Baidu Robotaxi Outage in Wuhan Strands Passengers in Traffic

A large-scale outage of Baidu’s Apollo Go robotaxis in Wuhan has left passengers stranded in moving traffic, after more than 100 autonomous vehicles reportedly froze in place across the central Chinese city.

Robotaxis Halt Across Wuhan During Evening Commute

Published coverage indicates that the disruption began on the evening of March 31, 2026, as commuters were returning home across Wuhan, a city of nearly 14 million residents in central China. Reports describe multiple Apollo Go vehicles coming to sudden stops on elevated ring roads and major arteries, in some cases blocking fast lanes while conventional cars continued to pass on both sides.

Local police statements cited in international media point to a “system malfunction” affecting more than 100 robotaxis. Several outlets describe the event as a mass shutdown, with vehicles entering a failsafe mode and refusing to move, even when traffic conditions were clear. No fatalities or serious injuries have been reported, but minor collisions and significant disruption to traffic have been noted in some accounts.

Social media images and videos referenced by news organizations show the pale green Apollo Go robotaxis stationary in the middle of multilane roads, hazard lights flashing, while other vehicles edge around them. The scenes underline how even a short-lived outage in a tightly synchronized urban transport network can ripple outward, causing congestion and confusion well beyond the stalled cars themselves.

Wuhan has been one of Baidu’s flagship testing grounds for large-scale deployment of autonomous ride services, with hundreds of robotaxis operating across thousands of square kilometers. That status has now drawn additional scrutiny as observers examine how the outage unfolded in a city that had been promoted as a showcase for commercial driverless mobility.

Passengers Report Being Trapped for Up to Two Hours

Accounts compiled from Chinese social media and reprinted in outlets such as the Associated Press, Sky News, and technology publications describe passengers caught off guard as their robotaxis abruptly came to a halt. Riders reported system messages appearing on in-car screens noting a driving system malfunction and indicating that staff were expected to arrive within minutes.

In practice, several passengers recounted waiting far longer than the initial estimates. Some reported being stuck for 30 to 90 minutes before they were able to safely exit, while others remained inside immobilized vehicles for close to two hours on elevated expressways. Videos shared online and cited in coverage show riders repeatedly trying to reach customer service from the car’s tablet interface, with calls allegedly going unanswered or looping through assurances that specialists had been dispatched.

For passengers whose robotaxis stopped in live traffic, the decision of whether to stay inside or step out into moving lanes posed a genuine dilemma. Reports suggest that some riders chose to wait for assistance, while others exited after determining there was a gap in traffic. In several instances, police or road workers appear to have helped riders leave vehicles and guided them to safety at the roadside.

The incident highlights an often-overlooked dimension of autonomous transport: the passenger support systems that must operate when automated driving fails. In Wuhan, the combination of a sudden technical fault, dense nighttime traffic, and limited real-time communication left some riders describing their experience as being “trapped” in the middle of the road, despite having physical control over the door locks.

Technical “System Failure” Raises Questions Over Redundancy

Available information so far suggests that the root cause of the outage was a centralized technical failure impacting core systems used by Apollo Go vehicles. Police statements cited by multiple outlets refer to a system malfunction, while some Chinese-language reports mention customer service attributing the breakdown to network issues that interfered with the driving system.

Experts quoted in prior analyses of robotaxi operations have emphasized that commercial fleets rely not only on onboard sensors and software, but also on high-definition maps, cloud-based routing, and continuous connectivity. The scale and synchronicity of the Wuhan disruption, with scores of vehicles stopping within a narrow time window, has prompted speculation that a shared backend component failed in a way that triggered conservative safety protocols across the fleet.

From a safety engineering perspective, bringing vehicles to a controlled stop in the event of a serious system fault is generally regarded as the correct default. Yet the Wuhan episode illustrates that a “safe stop” in theory may translate into a risky situation in practice when it occurs in the middle of a fast lane on an elevated roadway. The lack of human drivers to improvise or quickly maneuver to the shoulder removed a traditional safeguard that conventional taxis or ride-hailing vehicles might provide.

Regulators and industry observers are now likely to closely examine how Apollo Go’s failover and fallback behaviors are defined, including how vehicles decide where to stop, how passengers are instructed, and what contingencies exist if connectivity to remote assistance is degraded. The answers will shape public and governmental confidence in further expansion of driverless services, both in China and in overseas markets where Baidu has announced ambitious growth plans.

Impact on Baidu’s Expanding Robotaxi Ambitions

Baidu has invested heavily in positioning Apollo Go as one of the world’s largest robotaxi services, with operations in numerous Chinese cities and pilot schemes in the Middle East and planned deployments in Europe. Company filings and prior public statements show rapid growth in ride volumes, with Wuhan highlighted as a major hub and early pilot zone.

The Wuhan outage arrives at a sensitive moment for the broader autonomous driving sector. In recent years, high-profile incidents involving driverless fleets in the United States and China have intensified scrutiny of safety records, incident reporting, and the relationship between private technology firms and public authorities that regulate streets and highways. Against this backdrop, a single evening of disruptions affecting more than 100 vehicles in a major city can quickly become a global reference point in debates over the reliability of self-driving cars.

Financial and technology media report that Baidu’s share price dipped following news of the malfunction, reflecting investor concern about potential reputational damage and regulatory headwinds. Analysts note that large-scale outages raise questions not only about safety, but also about the robustness of the business model if fleets can be taken offline by a single technical fault.

Internationally, partners and prospective customers in cities where Apollo Go is seeking to expand may now demand more granular information on how the Wuhan incident unfolded, what mitigations are being added, and how future outages will be handled. The answers could influence timelines for new pilot programs and may inspire rival operators to highlight their own redundancy strategies and crisis response protocols.

Travel and Urban Mobility Implications for Wuhan

For residents and visitors in Wuhan, the immediate effect of the outage was a period of localized traffic chaos, compounded by confusion among drivers encountering motionless vehicles in active lanes. Reports describe delays on ring roads and elevated highways as emergency responders and towing services worked to clear stalled robotaxis and manage traffic flow.

In the short term, the incident may cause passengers to reconsider their choice of transport, particularly during peak hours or on routes that rely heavily on autonomous services. Travel and mobility analysts note that large Chinese cities have actively promoted robotaxis as a way to ease congestion and provide flexible, on-demand mobility. Events like the Wuhan outage challenge that narrative by demonstrating how quickly automated systems can themselves become a source of gridlock.

The episode is also likely to feed into ongoing discussions about how cities integrate autonomous vehicles into broader transport plans. Questions include whether robotaxis should be restricted from certain elevated or high-speed corridors until reliability metrics improve, how to ensure rapid physical response when driverless vehicles stall, and what role public communication plays in guiding passengers during system failures.

For travelers watching developments from abroad, Wuhan’s robotaxi outage serves as a reminder that the convenience of app-based autonomous rides comes with new types of risk. As more destinations experiment with self-driving shuttles and taxis, transport planners and technology providers will be under growing pressure to demonstrate that behind-the-scenes systems are as resilient as the sleek vehicles visible on city streets.