沙利文发布《2026年中国汽车AI云市场追踪报告（26H1）》
Frost & Sullivan Releases the "China Automotive AI Cloud Market Tracking Report (2026H1)"
汽车行业用云的核心驱动力，正由“业务系统数字化”转向“AI模型持续开发与运营”。随着高阶辅助驾驶、大模型座舱与工业AI进入量产与规模化阶段，云端开始承载数据治理、模型训练、仿真评测、推理部署与模型持续迭代等新工作负载，汽车云的价值边界由传统IT基础设施延伸至AI生产基础设施。规模上，中国汽车AI云市场由2025年的122.1亿元，预计2029年将达到753.0亿元，2025年至2029年复合增长率为57.5%。
汽车云竞争格局依然呈现整体集中的特征。汽车AI云市场集中度较高，阿里云以43.0%的份额领先，腾讯云、华为云紧随其后，前五大厂商合计约占据80.8%的市场。分场景看，自动驾驶场景由阿里云、百度智能云与华为云领跑；智能座舱场景由阿里云、腾讯云与火山引擎位列前三；工业提效场景呈现阿里云与华为云"两强引领"格局。三类场景在AI工作负载结构与商业逻辑上存在本质差异，单一能力难以覆盖全部场景，厂商优势因此高度场景化。
展望未来三到五年，汽车AI云的价值增长将从"训练模型"扩展至"持续运行和运营模型"，价值池由基础智算资源逐步迁移至训推一体、AI工程平台及MaaS与Agent持续运营，AI PaaS与MaaS成为增量核心。行业关键约束也已从算力供给转向数据可用性、成本经济性与合规能力。
The core driver of cloud adoption in the automotive industry is shifting from the digitalization of business systems toward the continuous development and operation of AI models. As advanced assisted driving, large-model cockpits and industrial AI move into mass production and scaled deployment, the cloud increasingly carries data governance, model training, simulation and evaluation, inference deployment and continuous model iteration, extending the value boundary of automotive cloud from traditional IT infrastructure to AI production infrastructure. In terms of scale, the China automotive AI cloud market stood at RMB 12.21 billion in 2025 and is expected to reach RMB 75.30 billion in 2029, representing a CAGR of 57.5% between 2025 and 2029.
The competitive landscape of automotive cloud remains concentrated overall. Concentration in the automotive AI cloud market is relatively high, with Alibaba Cloud leading at a 43.0% share, followed by Tencent Cloud and Huawei Cloud; the top five vendors together account for approximately 80.8% of the market. By scenario, autonomous driving is led by Alibaba Cloud, Baidu AI Cloud and Huawei Cloud; the intelligent cockpit is led by Alibaba Cloud, Tencent Cloud and Volcano Engine; and industrial efficiency shows a "two-leader" pattern led by Alibaba Cloud and Huawei Cloud. Given the fundamental differences in AI workload structure and business logic across the three scenarios, no single capability can cover all of them, so vendor advantages remain highly scenario-specific.
Looking ahead over the next three to five years, value growth in automotive AI cloud will extend from "training models" to "continuously running and operating them." The value pool is migrating from underlying intelligent computing resources toward integrated training and inference, AI engineering platforms, and MaaS and Agent-based continuous operation, making AI PaaS and MaaS the core of incremental growth. The industry's key constraint has also shifted from computing supply to data availability, cost economics, and compliance capability.
点击以下链接，查看节选版 报告：
如果您对中国汽车AI云市场有进一步的研究需求，请联系我们：
沙利文 李女士
E-mail: livia.li@frostchina.com
