Baidu's CFO, Haijian He, elaborated on the company's successful transition into a full-stack AI player, emphasizing that AI-related revenue now constitutes over half of its general business. In Q1 2026, AI-powered business accounted for 52% of revenue, hitting $1.89 billion (RMB 13.6 billion), a 49% year-over-year increase. This growth is primarily fueled by AI Cloud infrastructure, which saw revenue reach $1.22 billion (RMB 8.8 billion), an impressive 79% jump compared to the previous year. The company sees its proprietary Kunlunxin AI chips and comprehensive AI capabilities as key differentiators, allowing for cost optimization and improved customer mix, which supports margin expansion.
He further explained that Baidu's GPU cloud business is structurally higher margin than traditional CPU cloud, driven by robust demand, a tight supply chain, high technical barriers, and pricing power. The company's GPU cloud revenue surged by 184% year-over-year, significantly outpacing the broader market. This strong performance is attributed to the increasing enterprise demand for AI infrastructure, particularly for inference workloads, indicating that customers are moving beyond just training models and are actively deploying AI across their operations at an accelerating pace.
Deng Shen, President of Baidu's AI Cloud Group, noted that customers prioritize stability and availability over mere cost, seeking proven solutions that can scale, are compatible with mainstream models, and offer efficient cluster deployments. Baidu's full-stack offering, which includes cloud capacity, Qianfan services (model access), and Kunlunxin chip sales, effectively addresses these needs. The company also anticipates that its robotaxi business will evolve into a more convenient and integrated part of the public transportation ecosystem over time.
