As China's car market shifts from first-time purchases to replacement demand, artificial intelligence is poised to overhaul how automakers market their vehicles, according to a senior industry official speaking at the 2026 TEDA Automotive Forum on September 18.
Li Wenjun, general manager of the digital development department at the China Internet Association and deputy general manager of China Internet Research Institute, delivered a keynote speech titled "AI-Era Auto Marketing: Breaking Data Dilemmas, Empowering Full-Domain Growth." She argued that the domestic auto market has entered a new phase of stock replacement, requiring marketing logic to shift from traffic acquisition to value co-creation, according to Gasgoo.
Three structural shifts
Li identified three major structural changes reshaping the market: replacement purchases have overtaken first-time purchases as the core growth driver; car-buying decisions have moved significantly online; and consumers increasingly value full-scenario intelligent vehicle experiences. Regional market divergence is also becoming more pronounced, making refined, segmented operations essential.
She highlighted four persistent data challenges in auto marketing: user data is fragmented and disconnected; while observable user behavior exists, consumer motivations are hard to discern; user needs evolve dynamically with the external environment; and there is often a gap between users' stated preferences and their actual purchase choices. Traditional static labeling models, she said, are no longer sufficient in the stock market era, necessitating dynamic, human-centric insights based on multi-source data and motivation models.
From static labels to dynamic insights
To achieve deep understanding of individual users, Li cautioned against simply merging data tables. Instead, she proposed building a unified data foundation that ensures semantic consistency, temporal alignment, and confidence labeling, then constructing a closed-loop flywheel of perception, cognition, action, and sedimentation to continuously refine models. This foundation enables an AI lead-scoring engine that moves beyond static labels, integrating multiple signals to dynamically grade purchase probability, buying windows, and customer value, allowing sales teams to focus on high-intent prospects.
Li also presented a digital consumer modeling system that uses ten layers of structured constraints to reconstruct individual decision logic, enabling quantitative assessment across four dimensions: intention, risk, value, and strategy. This system can distinguish users who share the same labels but have fundamentally different decision-making processes. Derived tools include an AI focus group for early-stage product screening, simulation of market responses to different pricing and rights strategies, and a social simulator that tests market scenarios in virtual environments to reduce real-world marketing costs.
For Li, AI's impact on auto marketing is not about improving individual touchpoints but about systemic transformation. By leveraging digital consumer systems, lead-scoring engines, and social simulators, the industry can drive marketing intelligence, data assetization, and industrial collaboration. Future competition, she concluded, will not be about single-point technology superiority but about the coordinated upgrade of the entire industry ecosystem.
