News
Industry Dynamics | Digital Intelligence Industry

1.Release of China's First AI Agent Standard for Software Development

The China Academy of Information and Communications Technology, in collaboration with over 20 leading enterprises including Industrial and Commercial Bank of China, Baidu, Tencent, Alibaba, and Huawei, officially released the "Technical and Application Requirements for Software Engineering Intelligent Agents - Part 1: Development Agents" (Standard No. AIIA/T 0219-2025). This standard establishes a comprehensive technical specification for software development intelligent agents across two dimensions: technical capabilities and service capabilities, marking a significant step forward in the standardization process of AI agents in China.

2.Eight Ministries and Commissions Issue the "Special Action Plan for Accelerating the Development of Digital Intelligence Supply Chains"

The Ministry of Commerce, National Development and Reform Commission, Ministry of Education, Ministry of Industry and Information Technology, Ministry of Transport, Ministry of Agriculture and Rural Affairs, State Administration of Taxation, and National Data Administration jointly issued the "Special Action Plan for Accelerating the Development of Digital Intelligence Supply Chains". The plan aims to deepen the innovation and application of supply chains, fully leverage the role of digital intelligence supply chains in improving the modern commerce and circulation system, effectively reducing logistics costs across society, and enhancing the competitiveness and resilience of industrial and supply chains.

The plan proposes promoting the deep integration of the real economy and the digital economy, leveraging the role of leading supply chain enterprises, and utilizing new technologies such as AI, IoT, and blockchain to advance the digitalization, intelligence, and visualization of supply chains on a "chain-by-chain" basis. This will enhance capabilities in demand forecasting, intelligent decision-making, risk perception, and self-repair, thereby improving supply chain operational efficiency and resilience, and providing support for reducing logistics costs, building a modern industrial system, and optimizing consumption supply levels.

By 2030, a replicable and promotable model for the construction and development of digital intelligence supply chains will be formed, a deeply embedded, smart, efficient, and autonomous digital intelligence supply chain system will be basically established in important industries and key areas, around 100 national digital intelligence supply chain leading enterprises will be cultivated, and the resilience and security level of China's industrial and supply chains will be further enhanced.

To achieve this, the plan outlines 10 main tasks, including cultivating leading enterprises in digital intelligence supply chains, promoting the digital transformation of supply chain management services, accelerating the opening up and cooperation of digital intelligence supply chains internationally, building digital intelligence supply chain control towers, innovating underlying technologies for digital intelligence supply chains, improving technical service standard systems, exploring and advancing supply chain data transactions, ensuring the security of supply chain data flows, establishing a coordinated promotion mechanism for supply chains, and strengthening the cultivation of talent in digital intelligence supply chains.


3. Releases Industry's First Industrial Large Model Centered on Supply Chain - Joy Industrial

On May 22nd, JD Industrial officially released Joy Industrial, the industry's first industrial large model centered on the supply chain, in Shanghai.

Joy Industrial introduces AI intelligent agents such as demand agents, operation agents, and customs agents to meet the needs of JD Industrial and upstream suppliers in the supply chain, as well as AI products like commodity experts and integration experts to serve downstream enterprise users. It also sets clear goals for building industrial large models in key vertical industries such as the automotive aftermarket, new energy vehicles, robot manufacturing, oil and gas, and power grids.

4.OpenAI Announces Construction of the World's Largest AI Data Center in the UAE, Considering Expansion to the Asia-Pacific Region


OpenAI has reached a strategic cooperation agreement with G42 in the UAE to jointly build the world's largest AI data center cluster in Abu Dhabi. This project is part of OpenAI's "Stargate" plan and marks its first overseas expansion.

The super data center covers an area of approximately 10 square miles (about 26 square kilometers) with a total power demand of 5GW, equivalent to the power generation capacity of five nuclear power plants. This scale far exceeds OpenAI's first "Stargate" project in Abilene, Texas, USA, which is expected to have a total power consumption of only 1.2GW. As a project partner, G42, an AI company in the UAE supported by the country's sovereign wealth fund, will provide significant support.

OpenAI will be the main user of the data center, planning to utilize 1GW of computing power resources, with the remaining capacity potentially open to other enterprises. The project will adopt a phased construction strategy, with the first phase building a 1GW small cluster, of which 200 megawatts are expected to be operational by 2026 as a demonstration project for subsequent expansions.

5.Tencent's Comprehensive AI Strategy Unveiled

At the "2025 Tencent Cloud AI Industry Application Summit", Tencent Cloud unveiled a new knowledge base product matrix:

Tencent Lexiang: Upgraded to Lexiang Knowledge Base for enterprise organizations, providing rich functions in knowledge integration and sedimentation, knowledge updates, permission management, and AI Q&A to help enterprises effectively integrate and utilize private domain knowledge and improve the efficiency of internal knowledge circulation.

Tencent ima: An AI workstation for individuals and teams, integrating Hunyuan and DeepSeek large models, supporting the construction of personal knowledge bases, initiating AI Q&A, quickly interpreting materials, and generating content in notes. Creators can also publicly publish professional content through "Knowledge Number", which the system associates with similar Q&A raised by ima users to enhance the quality and influence of vertical Q&A. 

Tencent Qidian Marketing Cloud Intelligent Agent: Officially released during the main forum, it automates tasks such as audience extraction, portrait analysis, product optimization, marketing content production, and activity review based on large model reasoning, multi-agent architectures, marketing knowledge bases, and MCP technologies.

Tencent Cloud Zhijian Computing Series Products: Comprehensively upgraded in three directions: efficiency, reliability, and usability. Released the first Serverless platform in China supporting cross-card GPU scheduling; created a more cost-effective multi-machine interconnection network vRDMA; and combined products and solutions such as Xingmai Network, AI server inspection systems, cloud storage data万象 and GooseFS, vector databases, intelligent logging services, and large model security to provide a solid infrastructure for large models and applications.

Tencent Cloud CodeBuddy Craft Software Development Intelligent Agent: Developers can describe their needs in natural language, and Craft will automatically disassemble tasks, design modules, generate code, and self-correct errors, making "one-sentence application development" a reality.

Continuous Evolution of Tencent Cloud Products in Toolchains, Office Collaboration, Data Analysis, and Other Fields: Including Tencent Meetings, Tencent Cloud BI Intelligent Assistant ChatBI, Tencent Questionnaire, Tencent Electronic Signature, Tencent Documents, etc., further promoting AI as a natural part of enterprise daily toolchains.

6. TCL and Alibaba Cloud Reach Full-Stack AI Strategic Cooperation to Build Vertical Domain Professional Large Models

 On May 21st, TCL and Alibaba Cloud announced a full-stack AI strategic cooperation. Both parties will focus on semiconductor displays, smart terminals, and other fields, jointly building vertical domain professional large models based on Alibaba Cloud's "Cloud + AI" core capabilities.

 Research and Development Efficiency Improvement: TCL and Alibaba Cloud will continue to iterate and optimize the semiconductor display expert large model Xingzhi X-Intelligence based on models such as Qwen3, Qwen-VL, and QwQ. Xingzhi will summarize professional knowledge from reports and papers in the semiconductor field, providing technical references for R&D personnel and improving R&D efficiency.

 Product Innovation: Leiniao Innovation and Alibaba Cloud have reached an exclusive strategic cooperation. Thongyi series large models provide exclusive customized technical support for Leiniao AI and AR glasses.

 Going Global: Leiniao Technology's TV content operations, search, promotion, advertising, and other core businesses in Europe run on Alibaba Cloud. Meanwhile, Leiniao Innovation's AI glasses are also going global based on Alibaba Cloud's global infrastructure, using Alibaba Cloud's underlying computing power for spatial video computing and vertical agent construction, multi-modal large model fine-tuning

7.Baidu Wenxin Large Model X1 Turbo Wins Highest Rating Certificate from China Academy of Information and Communications Technology

At Baidu AI Day on May 20th, the China Academy of Information and Communications Technology announced the evaluation results of large model reasoning capabilities.

Baidu Wenxin Large Model X1 Turbo achieved the highest current rating of "4+ level" in 24 capability assessments, with 16 items scoring 5 points, becoming the first domestic large model to pass this evaluation.

 Wenxin X1 Turbo received full marks in technical capabilities such as logical reasoning, code reasoning, reasoning effect optimization, and application capabilities such as tool support and security reliability.