Major Chinese tech companies like Alibaba, Tencent, and Huawei are making big strides in developing agentic AI systems. These are AI tools that can perform complex, multi-step tasks on their own, without needing human instructions. They are also focusing on tailoring these systems to specific industries and workflows, such as finance, manufacturing, and logistics.

Alibaba’s Open-Source Approach to Industry-Specific AI

Alibaba is centered around its Qwen AI model family, a set of large language models with multilingual capabilities that are open source. These models serve as the foundation for AI services and autonomous agent platforms offered through Alibaba Cloud. The company has made its agent development tools and vector database services publicly available, allowing developers to adapt and build their own autonomous agents.

This approach positions Qwen models as a platform for creating industry-specific solutions. Alibaba has launched applications built on these models, like the Qwen App, which has gained a large user base since its beta release. The app connects autonomous AI tasks with Alibaba’s broader ecosystem, including commerce and payments. Additionally, Alibaba has released an open-source agent framework called Qwen-Agent, encouraging third-party developers to build autonomous systems, much like Western projects such as Microsoft’s AutoGen or OpenAI’s Swarm.

Tencent and Huawei’s Industry-Focused AI Initiatives

Tencent has also released an open-source agent framework named Youtu-Agent. While Tencent’s cloud presence remains smaller outside China, they are investing heavily in scenario-based AI tools and SaaS applications that target enterprise needs. These are designed to help businesses automate tasks and improve workflows.

Huawei, on the other hand, combines model development, infrastructure, and industry-specific agent frameworks to attract users worldwide. Its Huawei Cloud division has developed a ‘supernode’ architecture that supports large-scale cognitive models and workflow orchestration. Huawei’s Pangu family of models are embedded with AI agents tailored for telecoms, utilities, creative industries, and manufacturing. Early deployments include sectors like network optimization, manufacturing, and energy, where AI agents help with predictive maintenance and resource management with minimal human input.

Despite these efforts, most Chinese agentic AI applications are currently visible within China. Projects like OpenClaw, created outside the Chinese ecosystem, have been integrated into workplace tools such as Alibaba’s DingTalk and Tencent’s WeCom. These tools are used for automating scheduling, coding, and managing developer workflows. While these integrations are popular among Chinese developers, they have not yet gained widespread global adoption.

Overall, Chinese hyperscalers are rapidly developing and open-sourcing tools for autonomous AI tailored to industry needs. Their focus on open frameworks and specialized solutions signals a competitive push into the enterprise AI market, both within China and increasingly on a global scale.

Inspired by

• https://www.artificialintelligence-news.com/news/chinese-hyperscalers-and-industry-specific-chinas-agentic-ai/

Sources

• chinastrategy.org