China’s brutal electric vehicle price war is entering a new phase, but it’s not bringing relief to manufacturers. As ByteDance and Alibaba rush to embed their AI technologies into vehicles, what started as a competitive edge is rapidly becoming table stakes. The shift reveals a troubling pattern: in China’s hyper-competitive EV market, even cutting-edge AI features are getting commoditized before they can deliver meaningful margins.

The conference rooms at Chinese EV startups have a new obsession, and it’s not battery range or charging speed anymore. It’s voice assistants, personalized recommendations, and AI-powered everything. But there’s a problem: everyone’s having the same conversation.

ByteDance, the company behind TikTok, has been pushing its Doubao AI assistant and Volcano Engine cloud infrastructure into vehicles at breakneck speed. Meanwhile, Alibaba is leveraging its cloud computing dominance to embed its own AI capabilities into dashboards across the industry. What sounds like healthy competition is actually revealing a deeper structural issue in China’s EV sector.

The playbook is familiar to anyone who watched China’s smartphone wars. A manufacturer announces a breakthrough feature – say, an AI assistant that can predict your destination based on calendar entries and traffic patterns. Within weeks, three competitors announce similar capabilities. Within months, it’s standard across the segment. The differentiation evaporates before the first units hit showrooms.

This time, the cycle is moving even faster. Chinese automakers, squeezed by razor-thin margins from years of price competition, saw AI as their salvation. Smart cockpits powered by large language models could justify premium pricing and create sticky ecosystems. That was the theory, anyway.

The reality is messier. ByteDance’s aggressive push into automotive through Volcano Engine has made advanced AI capabilities widely accessible. Any manufacturer with integration resources can now offer conversational AI, personalized content streaming, and predictive features. Alibaba’s parallel effort through its cloud division has the same democratizing effect. When everyone can license similar underlying technology, the innovation becomes a commodity.

Industry analysts point to broader implications beyond just features. The original EV price war forced manufacturers to cut costs so aggressively that many are selling vehicles at or below production cost, banking on future software revenues to make up the difference. But if AI features commoditize as quickly as they’re being deployed, that software revenue thesis looks shaky.

The pattern mirrors what happened with advanced driver assistance systems. What started as premium features in high-end models became standard equipment across price points within 18 months. Manufacturers who invested heavily in developing proprietary systems found themselves competing with suppliers offering similar capabilities to everyone.

For ByteDance and Alibaba, the automotive sector represents the next frontier after saturating the smartphone market. Both companies are betting that cars will become the primary computing platform for consumers during commutes and road trips. The competition isn’t really about helping EV makers differentiate – it’s about controlling the interface between drivers and digital services.

But this creates a fundamental tension. EV manufacturers need exclusive features to stand out in a crowded market. Tech platforms need scale and ubiquity to win the ecosystem war. Those objectives don’t align, and the manufacturers are learning this the hard way.

Some startups are already pivoting strategy, trying to build truly proprietary AI capabilities in-house rather than relying on third-party platforms. That’s expensive and risky in an industry already burning cash. Others are doubling down on integration and user experience, hoping that how AI features work matters more than what they can do. That’s a tough sell when the underlying models are essentially identical.

The commoditization also raises questions about data and privacy. As more vehicles rely on cloud-based AI from ByteDance and Alibaba, those platforms accumulate detailed information about driving patterns, destinations, and user preferences. That data could become more valuable than the vehicle sales themselves, but it’s unclear how much of that value flows back to automakers versus staying with the platform providers.

What’s particularly striking is the speed of this cycle. The EV price war took several years to reach its current brutal state. The AI features war seems to be compressing that timeline into months. Announcements that would have dominated headlines a year ago now barely register because three competitors announced something similar last week.

China’s EV industry is learning a harsh lesson about sustainable competitive advantages. The shift from price wars to AI features was supposed to move competition up the value chain, creating differentiation through software rather than just cheaper hardware. Instead, it’s revealing that in a market with abundant capital, fast-moving suppliers, and desperate manufacturers, even sophisticated AI capabilities can become commodities almost overnight. The winners in this environment won’t be the companies with the most features – they’ll be the ones who figure out how to build moats that actually hold water. For now, that remains an unsolved problem, and the cash burn continues.