- Published on
Baidu Intelligent Cloud reportedly splits platform product division: MaaS integrated into infrastructure, Agent estab...
- Authors

- Name
- DeepHorus
Baidu Intelligent Cloud has recently undergone a new round of organizational restructuring within its product and research divisions, aiming to focus more closely on its core artificial intelligence (AI) business. The original Platform Products Business Unit has been officially renamed the Intelligent Agent Business Unit, while its subordinate MaaS (Model as a Service) business, the Qianfan brand, and related product systems have been wholly transferred to the Infrastructure Business Unit. Additionally, the Data Platform Department has been renamed the Data Intelligence Department and incorporated into the Intelligent Application Business Unit, which was formerly known as the Application Products Business Unit.
Following this adjustment, Baidu Intelligent Cloud has formed three core business units. The Infrastructure Business Unit, led by Hou Zhenyu, primarily manages AI Infrastructure and Agent Infrastructure, undertaking the support for underlying computing power and platforms. The Intelligent Agent Business Unit, overseen by Yin Dawei, focuses on the Agent product line, including applications like Baidu Dazi, Miaoda, and Famou. The Intelligent Application Business Unit, headed by Ruan Yu, is responsible for Baidu Yijian, Hogee, Shengsuan, as well as industry solutions for transportation, finance, and other sectors. The three leaders all report directly to Shen Dou, Executive Vice President of Baidu Group and President of the Intelligent Cloud Business Group.
The background for this organizational change is the robust growth of Baidu's AI business. According to the company's second-quarter 2026 financial report released this month, total revenue reached 31.3 billion yuan, with general business revenue at 25.2 billion yuan. Notably, AI business revenue already accounts for 50% of the general business revenue and has maintained over half the share for two consecutive quarters. Revenue from GPU cloud within AI cloud infrastructure saw a significant year-on-year increase of 283%, marking the fourth consecutive quarter of triple-digit growth.
Management stated clearly in the earnings report that they will continue to increase investment in the AI field, driving the company's transformation from a traditional internet enterprise to an AI-first company. The sustained growth of the AI business is viewed as the core engine for the company's long-term development. This restructuring is precisely aimed at optimizing resource allocation to better support this strategic direction.
At the specific business level, Baidu has already initiated the integration of its internal office AI agent "dodo" with the "Baidu Dazi" team this month. All R&D personnel and resources related to dodo will be merged into the Baidu Dazi team, achieving complete unification of internal and external office AI agent products. This signifies that in the AI office track, Baidu is shifting from previous multi-pronged attempts to focusing its efforts on key breakthroughs, aiming to consolidate superior resources and fully participate in market competition.
The transfer of the MaaS business to the Infrastructure Business Unit underscores Baidu's strategic judgment of Model as a Service as a foundational capability. The independence of the Intelligent Agent Business Unit indicates the company's high regard for the prospects of Agent applications. The adjustment aims to connect the complete technical chain from infrastructure and model services to upper-layer intelligent agent applications, thereby enhancing overall collaborative efficiency.
Looking ahead, the integrated organizational structure is expected to accelerate Baidu's innovation and commercialization processes in the AI field. By focusing on Agent products and industry intelligent applications, Baidu Cloud is poised to form a clearer product matrix and competitive advantages in the intense race of cloud computing and AI, further consolidating the achievements of its AI-first transformation.