Gasgoo Munich-At the Chengdu Motor Show, "no quick-build cars" suddenly became a catchphrase. Traditional automakers like SAIC Volkswagen, Dongfeng Peugeot-Citroën, Jetta, and Beijing Hyundai have spoken up, emphasizing that while R&D can be accelerated, reliability verification cannot be compromised. Yet this stance hasn't won unanimous approval. On one side, skeptics question whether joint ventures—criticized for their sluggish pace in the shift to electrification and intelligence—are simply finding a more dignified excuse for their slowness. On the other, automakers advocating for platformization, simulation, and digitalization to slash development cycles face a different question: if cars launch faster, are early buyers just becoming test subjects? The result is an absurd paradox: build too slow, and you get criticized; build too fast, and you get criticized too. But the real debate isn't about whether a car should take 18 or 36 months to build. It's about this: what got cut? Why are joint ventures leading the charge against "quick-build" cars? During the Chengdu Motor Show on August 21, Li Jun, executive director of SAIC Volkswagen's marketing division, told reporters that some new models have slashed the timeline from project approval to mass production to 18 months or less. But, he warned, "if you save time in too many places, you're bound to face hidden risks down the road." Shortly after, Yu Yuefeng, deputy general manager of Dongfeng Peugeot-Citroën, echoed that sentiment: "We are not in the business of building rushed cars." Beijing Hyundai, Great Wall Motor, and SAIC-GM have also joined the public discussion. Jetta is taking a similar stand. As the brand's first pure electric model following its refresh, the Jetta M6 made its debut at the show and opened blind pre-orders. The company emphasized that the car endured "two winters and two summers," tested in extreme conditions from the freezing cold of Heihe to the scorching heat of Turpan. During the show, Jetta General Manager Gao Jiefang told reporters that he personally tests the M6 almost every week, offering feedback and checking adjustments the following week. In the early hours of the second day of the show, he and his team were back at it, testing the M6's driver-assist capabilities in low-light conditions. It's not hard to understand why traditional automakers are suddenly stressing "verification" en masse. For the past few years, the Chinese market has rewarded speed. Data from the China Association of Automobile Manufacturers shows that Chinese brands sold 9.138 million passenger vehicles in the first half of 2026, capturing 71.8% of the market. Joint ventures and foreign brands were left with just 28.2%. The gap in product offerings is even starker. According to incomplete statistics from Gasgoo, the domestic passenger market launched 116 new or redesigned models in the first half of this year. Domestic brands and startups accounted for 90 of those—or 77.6%—while joint ventures managed only 26. Underlying this split are two fundamentally different development systems. Chinese NEV makers typically have shorter decision-making chains, involve suppliers earlier, and run R&D, product planning, and manufacturing in parallel. Multinational automakers, meanwhile, often grapple with more complex global decision-making and supply chain structures. But not all of the "slowness" at joint ventures can be chalked up to inefficiency. A global model sold across multiple countries must navigate varying regulations, road conditions, climates, and user habits—all while ensuring consistency across different factories and supplier networks. So the real comparison isn't 18 months versus 36. It's about who can complete sufficient verification with less wasted time, given products of equal complexity. This is also where the "anti-quick-build" argument is most easily misunderstood. Global adaptation, regulatory certification, and supply chain validation take time. But lengthy approvals, departmental silos, and inefficient decision-making cannot be justified simply because "cars are complex industrial products." What joint ventures really need to explain is which delays stem from rigor—and which from inefficiency. When building a car, what time simply cannot be saved? Breaking down the development cycle, it isn't actually that complicated. Approvals, waiting, and repetitive communication should be cut. Mature platforms can be reused. Software, crash tests, thermal management, and many driver-assist scenarios can be validated early via simulation. These are efficiencies gained through technological and managerial progress. The real controversy lies in verification in the physical world. Li Jun was blunt when explaining why SAIC Volkswagen insists on "two winters and two summers" at the Chengdu show: "The first time exposes all the problems; the second time, you fix them completely, then test and verify again." In his view, metal fatigue, rubber aging, battery cycling, and overall vehicle durability simply take time. Wu Gansha, co-founder, chairman, and CEO of UISEE, echoed a similar sentiment in a recent conversation. He emphasized that extreme cold, blizzards, sandstorms, high heat, and low visibility can all compromise sensors, vehicle control, and even electrical systems. The trouble is, some extreme weather data cannot be fully replicated in a lab. So a problem unsolved one winter might have to wait until the next year's weather returns for further verification. All of this points to the issue most easily overlooked in the "quick-build" debate. Testing isn't just about racking up miles; it's about leaving time for failure. If winter testing reveals a flaw, you need a second round of verification after fixes. If durability testing shows structural anomalies, you need to retest after replacing parts. So the truly dangerous kind of "fast" isn't necessarily driving 5,000 kilometers less. It's when the project schedule is so compressed that there's no buffer: a major issue surfaces in the final test, but the launch deadline is locked in, leaving no time for a second round. This is why the question of "how much reliability testing is enough" is back on the regulatory radar. In January 2026, the Ministry of Industry and Information Technology revised requirements for vehicle market access, mandating that traditional cars undergo at least 30,000 kilometers of reliability testing. Under current rules, the total for electric vehicles is 15,000 kilometers. Meanwhile, the National Automotive Standardization Technical Committee has started revising standards, proposing to raise the reliability testing threshold for pure electric, plug-in hybrid, and fuel cell vehicles to a unified minimum of 30,000 kilometers. In July, the MIIT's Department of Equipment Industry also convened a meeting with key automakers, ordering them to review issues related to production consistency, reliability, durability, and the verification of new technologies. The regulator's signal is clear: cars can be developed faster, but that doesn't make verification any less critical. Simulation vs. Road Testing: The Real Debate Isn't About the Ratio This is the point Lu Fang has repeatedly stressed while responding to the "quick-build" controversy. "Short development time doesn't mean a car is a quick-build," he argued on August 12. "The only yardstick for whether a car is 'rushed' isn't time, but whether it has completed all necessary verification processes." Two days later, he clarified that simulation and real-world road testing are "not substitutes, but complementary, symbiotic, and iteratively linked." His description was vivid: simulation conducts "large-scale mock exams," while road tests conduct "precision field tests." Problems found on the road become new questions for simulation, creating a closed loop. The logic itself is sound. Extreme scenarios for driver assistance, software logic, and electronic systems are well-suited for simulation. Real roads cannot produce dangerous situations like cars cutting in or pedestrians darting out every day, but virtual environments can test them repeatedly. 51Sim end-to-end intelligent driving simulation solution; Image source: 51Sim official website But the closer you get to long-term physical changes, the more critical physical verification becomes. Will connectors fail after repeated thermal cycling? Will seals degrade? Will suspensions show anomalies after long-term impact? These are questions hard to answer simply by adding virtual miles. So the true value of simulation isn't driving fewer physical test miles. It's filtering out known issues early, reserving expensive and time-consuming real-world testing for the hardest-to-simulate problems. Therefore, whether the simulation ratio is 60%, 80%, or 90% is actually beside the point. The real question is whether that remaining portion of physical testing is tackling the toughest challenges. First-Time Buyers Should Not Be the Final Check The "quick-build" debate is intensifying partly because the market window for automakers is shrinking rapidly. According to statistics from Li Yanwei, a member of the China Automobile Dealers Association's expert committee, about 165 new models launched domestically in the first half of 2026. In May of this year, nearly three-quarters of NEV sales came from models launched or refreshed less than six months ago. Launch six months late, and you might be facing a different set of rivals. Launch a year late, and a once-leading feature could be standard equipment. This means engineering development is naturally squeezed by commercial rhythms. The problem is, a car is not a smartphone. Software can be updated continuously via OTA, but hardware cannot indefinitely follow the internet logic of "launch first, fix later." A voice assistant getting smarter in six months is product evolution. Adding new driver-assist features is also evolution. But if basic hardware reliability issues surface after delivery—only to be fixed through part replacements, mid-cycle refreshes, or recalls—that's a different story. Of course, a post-launch failure doesn't automatically prove insufficient verification. Image source: XPENG For instance, in July this year, some XPENG X9 owners experienced air spring failures in extreme heat. XPENG apologized and extended the warranty on the front air spring system to 8 years or 160,000 kilometers. Whether an issue stems from early verification, supplier consistency, production batches, or extremely low-probability failure requires specific analysis. But such events at least remind the industry: the real world will always throw up problems the lab didn't cover. A more reasonable boundary, therefore, is this: users can help automakers discover long-tail issues, but they should not be part of basic reliability verification. If a car is delivered to hit a market window before the verification loop is closed, and early user feedback is packaged as "co-creation," or if problems that should have been solved in engineering are explained away as "rapid iteration," then co-creation becomes nothing more than shifting trial-and-error costs onto consumers. This is the "quick-build" risk consumers truly fear—not the 18-month timeline itself. It's the question of whether they're buying a fully verified product, or a car that is sellable but not quite finished. Back at the Chengdu Motor Show, this debate has no clear winner. Traditional automakers are right to emphasize long-term verification, but they cannot explain away slow EV transitions or sluggish intelligent features as mere respect for industrial laws. Lu Fang is also right that platformization, digitalization, and simulation boost efficiency. But "faster" must come with an explanation: where exactly did the speed come from? Cutting approvals, wait times, and redundant development is efficiency. Shortening cycles through platform reuse and simulation is engineering progress. What truly cannot be compressed are critical verification—and the chance to start over when problems are found. Image source: Liu Jiaming's Weibo So for the next stage of China's auto industry, the real proof isn't whether a car can be built in 18 months. It's whether the car delivered after those 18 months is fully verified, or still waiting for the consumer to finish the job. Truly excellent automakers need not take pride in being "slow," nor do they need to boast about being "fast." The ultimate metric in this "quick-build" debate should be this: who can minimize wasted time while maximizing the integrity of essential verification.
Liked by Jackson Bouchard, Amélie TurnerLiked by Jackson Bouchard, Amélie Turner
Liked by Jackson Bouchard, Amélie Turner
