China is aiming to move autonomous driving beyond limited trials and towards mass deployment by 2030, pairing the target with a new generation of mandatory safety standards due to take effect from 2027.
China has set out a new roadmap for the large-scale deployment of autonomous vehicles by the end of the decade, as the country pushes to strengthen its position in smart electric vehicles and next-generation automotive technology.
According to Reuters, China’s Industry Ministry is targeting mass deployment of self-driving vehicles by 2030, backed by measures covering technological development, regulation and closer industry coordination.
The target matters because it points to a shift in emphasis. Autonomous driving has spent years being demonstrated through pilot programmes, limited robotaxi services and increasingly capable driver-assistance systems. China is now setting policy around the point at which those technologies are expected to operate at much greater scale.
China wants autonomous driving to move beyond the pilot stage
The 2030 target is part of a broader effort to develop China’s intelligent and connected-vehicle industry rather than a single launch date for fully driverless cars.
That distinction is important. “Autonomous driving” covers several levels of automation, from systems that still require continuous driver supervision through to vehicles capable of handling the driving task themselves within defined conditions.
China has already started putting more formal rules around those different levels. In August, the Ministry of Industry and Information Technology published the mandatory GB 44721—2026 autonomous-driving safety standard, which is scheduled to take effect on 1 July 2027.
The standard applies to vehicles fitted with Level 3 conditional automation and Level 4 highly automated driving systems, excluding automated parking. It establishes requirements covering manufacturer risk management, testing, system behaviour, human-machine interaction and the way drivers are informed about what an automated system can and cannot do.
Safety rules are arriving before mass deployment
One of the strongest signals in China’s approach is that regulation is being developed before the 2030 deployment target is reached.
Under GB 44721—2026, manufacturers are expected to build safety management into the full lifecycle of an automated-driving system, including development, production and post-deployment operation. Systems must also undergo simulation, proving-ground and road testing, while Level 3 systems need driver takeover monitoring and clear status information for users.
The ministry says an automated-driving system’s safety performance should at least match that of a qualified, attentive driver carrying out the same dynamic driving task, while also providing a minimum-risk strategy when the system can no longer continue normally.
A separate mandatory standard for combined driver-assistance systems was published in June and is due to take effect on 1 January 2027. The GB 47955—2026 standard covers systems that control both steering and vehicle speed while still requiring the driver to monitor traffic and remain responsible for the vehicle.
MIIT says 70% of new passenger vehicles sold in China during 2026 have been equipped with combined driver-assistance functions, with more than 30% offering navigation-assisted driving features. That makes the distinction between assisted and genuinely automated driving increasingly important for both manufacturers and drivers.
Why China’s 2030 target matters globally
This is not a suggestion that China’s rules will simply become UK, European or US rules. Each market has its own regulatory system, road environment and timetable for allowing higher levels of automation.
But China’s scale as a vehicle and technology manufacturing centre means a move towards mass deployment can influence where carmakers, sensor suppliers, chip manufacturers and software developers direct investment. China is also part of the international rule-making effort: in June 2026, UNECE’s WP.29 adopted a new global framework for automated-driving systems backed by major automotive markets including China, the EU, UK, US, Canada and Japan. UNECE says the framework introduces harmonised safety requirements covering testing, safety cases and in-service monitoring.
Other markets are already moving along their own paths. In the UK, Uber and Wayve are preparing autonomous EV services in London, while in the US Tesla has begun carrying passengers in its Cybercab robotaxis in Austin.
Those projects show how differently autonomous driving can develop from one market to another. Some programmes are built around tightly controlled commercial services, while China’s latest roadmap is aimed at creating the conditions for much broader industry deployment.
Official video: The footage below is from Pony.ai and shows the company’s Auto China 2026 push to commercialise Level 4 robotaxis and other autonomous vehicles at scale.
The real test is deployment, not the target
A 2030 target does not guarantee that autonomous cars will suddenly become commonplace across China four years from now. Safety validation, regulation, infrastructure, manufacturer readiness and public confidence still determine how quickly individual systems can move from testing into everyday use.
What makes the latest roadmap significant is the combination of scale and regulation. China is not only signalling that it wants autonomous vehicles deployed in large numbers; it is also putting mandatory safety baselines in place before that expansion takes place.
Tech Torque Verdict
China’s 2030 autonomous-driving target is ambitious, but the more important development may be what happens before then. Mandatory standards for driver assistance, Level 3 and Level 4 automation create a clearer line between systems that assist a driver and systems designed to take over the driving task.
If China can combine that regulatory framework with reliable large-scale deployment, the effects are unlikely to stop at its borders. Carmakers and technology suppliers operating globally will be watching closely — not simply to see how many autonomous vehicles reach the road, but whether the safety framework can keep pace with them.
