Shenzhen-based DeepRoute.ai now claims more than 300,000 passenger vehicles in China are running its assisted-driving system, a figure that has roughly tripled since the company reported passing the 100,000 mark earlier in 2025. If accurate, the milestone would make DeepRoute.ai one of the largest third-party suppliers of navigation-on-autopilot (NOA) technology in the world’s biggest auto market.
The number, disclosed through official company channels, lands at a moment when third-party ADAS suppliers are locked in a high-stakes race to win contracts from automakers that lack the resources or desire to build full self-driving stacks in-house. For Chinese consumers, the practical upshot is that advanced highway and city driving-assistance features are showing up in increasingly affordable cars, not just premium models.
A fast-moving milestone trail
DeepRoute.ai’s growth claims follow a documented sequence of its own press releases. The company said it had equipped more than 100,000 passenger cars when it unveiled its mass-production platform at the IAA 2025 event. That count rose to roughly 150,000 in a subsequent disclosure tied to announced plans to launch robotaxi operations using consumer-grade production vehicles by the end of 2025. As of late April 2026, there is no public confirmation that those robotaxi services have begun commercial operation; the linked press release describes a plan, not a launched service. By October 2025, the company said it was on track for 200,000 vehicles by year-end and claimed 40% of the third-party NOA market that month.
The jump from a projected 200,000 to a stated 300,000-plus represents a significant overshoot of the company’s own earlier forecasts. DeepRoute.ai has also floated an ambition to reach one million deployments, though it has not attached a specific deadline to that target.
All figures are self-reported
Every number in the chain originates from DeepRoute.ai’s own statements distributed via PR Newswire. No automaker partner, regulatory body, or independent research firm has publicly confirmed any of the totals. China’s Ministry of Industry and Information Technology tracks vehicle production and sales, but its public reports do not break down installations by specific ADAS supplier. Until third-party data sources begin tracking NOA installations more systematically, the company’s press releases remain both the primary window into its progress and a reminder of the limits of self-reported data.
The absence of independent corroboration is notable. Analysts at firms such as CINNOResearch and IDC, which publish regular reports on China’s intelligent-driving market, have not publicly cited DeepRoute.ai’s 300,000 figure or offered their own estimate of the company’s installed base. Without that external check, readers should weigh the milestone as a corporate claim rather than a verified industry statistic.
The 40% market-share claim deserves particular scrutiny. Market share requires a defined denominator, and DeepRoute.ai did not specify whether “third-party NOA market” covers all vehicles sold with such systems in China during October 2025 or a narrower slice filtered by price band or vehicle class. Without that methodology, the figure is best understood as the company’s own characterization of its competitive position, not an independently measured statistic.
Key questions the data does not answer
Which automakers are involved? DeepRoute.ai has not disclosed which OEM partners or vehicle models account for the 300,000-plus total. The company is known to have worked with Chery and other domestic brands, but aggregated figures obscure whether the deployment base is spread across many manufacturers or concentrated in one or two partnerships. A diversified roster would signal broad market acceptance; heavy reliance on a single brand would tell a different story.
What does “equipped” actually mean? The distinction between a car that shipped with DeepRoute.ai hardware and one whose driver has activated and regularly uses NOA features matters enormously. Factory installation counts and real-world usage are not the same thing, and the company has not clarified which definition it applies.
How many vehicles are private cars versus planned robotaxi fleet units? DeepRoute.ai’s announced strategy calls for the same hardware and software stack to serve both private owners and commercial fleets. However, because the company has not confirmed that its robotaxi service is operational, it is unclear whether any of the 300,000-plus vehicles are currently running in commercial ride-hailing mode. Without a breakdown, it is difficult to know how much of the company’s footprint reflects everyday private driving versus vehicles earmarked for future commercial use.
Where DeepRoute.ai fits in a crowded field
Third-party NOA suppliers occupy a specific niche in China’s autonomous-driving ecosystem. They compete not only against each other but also against automakers that build proprietary systems in-house. Companies like Huawei, whose Advanced Driving System powers vehicles from Changan, BAIC, and Seres, and Momenta, which supplies NOA technology to brands including SAIC and Toyota’s Chinese joint ventures, are fighting for the same contracts. Horizon Robotics, which went public in Hong Kong in late 2024, supplies the underlying computing chips that many of these systems run on, adding another layer of competition.
A third-party provider reaching 300,000 vehicles, if verified, would suggest that a meaningful number of automakers have decided that outsourcing NOA capability is faster and cheaper than building it themselves. For buyers, that trend could accelerate the arrival of highway and urban driving assistance in vehicles priced well below the premium segment where such features first appeared.
Regulation could reshape the timeline
Chinese authorities have broadly encouraged experimentation with autonomous and assisted driving, but the regulatory environment is not static. Rules governing automotive data security, high-definition mapping, and over-the-air software updates have tightened in recent years. Any future requirement for explicit feature-level approvals or stricter data-localization mandates could slow the pace at which DeepRoute.ai and its rivals scale deployments. The company’s public statements have not addressed how compliance obligations might affect its expansion plans.
What the 300,000 figure actually tells us about third-party NOA adoption in China
The most grounded reading of DeepRoute.ai’s claim is that the company has moved well beyond pilot projects into mass deployment. The numbers are internally consistent across multiple disclosures, and the growth pattern aligns with broader industry trends: Chinese consumers are adopting ADAS features faster than buyers in most other markets, and automakers are under intense pressure to offer competitive assisted-driving packages.
But the exact magnitude and composition of that scale remain opaque. How DeepRoute.ai’s footprint compares with rival suppliers and in-house automaker systems is still an open question. Consumers and investors should treat the 300,000-plus figure as a credible signal of meaningful traction while keeping in mind that independent verification has yet to catch up with the company’s own accounting.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
