Huawei Cloud Enters AI Agent Arena as Tech Giants Compete in "Lobster" Development

Deep News
Yesterday

Following Tencent and Alibaba, Huawei has officially announced its entry into the "lobster farming" competition. At the recent Huawei China Partners Conference 2026, Zhou Yuefeng, Senior Vice President of Huawei and CEO of Huawei Cloud, revealed that in the second half of this year, Huawei Cloud will launch a series of industry and scenario-specific "lobsters" based on the AgentArts one-stop enterprise-grade intelligent agent development platform. These will include OfficeClaw for office tasks, CodeClaw for programming, and industry-specific "lobsters" for marketing and customer service, further enhancing the enterprise AI application ecosystem. The term "lobster" is an industry nickname for AI agents, referring to their powerful "claws" capable of executing specific tasks. After the Spring Festival in 2026, this sector heated up rapidly, with tech giants like Tencent and Alibaba taking the lead. Huawei's entry now marks a new phase in the competition surrounding intelligent agents.

Tech giants are competing in the "lobster" sector, with Huawei Cloud focusing on industry-specific agents. Tencent has been one of the most active internet companies in this wave of "lobster farming." On March 22, WeChat announced the official launch of the WeChat ClawBot plugin, supporting integration with OpenClaw. Users updating WeChat to the latest version can find it in "Settings - Plugins," scan a QR code or copy a command to connect OpenClaw to WeChat. Once connected, users can quickly interact with their "lobsters" via WeChat chat for efficient task handling. Tencent has now established an intelligent agent matrix covering individual users, developers, and enterprise deployments.

Alibaba has demonstrated its commitment to the intelligent agent era through organizational restructuring. On March 16, Alibaba formally established the Alibaba Token Hub (ATH) business group, directly overseen by CEO Wu Yongming. During the post-earnings conference call, Wu Yongming announced a target of achieving over $100 billion in annual revenue from cloud and AI commercialization within five years, underpinned by a deep bet on exponential growth in token consumption driven by intelligent agents.

On March 20, at the Huawei China Partners Conference 2026, Zhou Yuefeng disclosed that Huawei Cloud will release a series of industry and scenario-oriented "lobsters" in the second half of the year. He emphasized that for AI to develop healthily and avoid becoming a bubble, it must serve as a tool to enhance enterprise productivity rather than merely providing emotional value. According to reports, AgentArts will be released for public testing in April, allowing developers and enterprises to use it free of charge on Huawei Cloud to build their own intelligent agents. In May, Huawei will also release an open-source version of the intelligent agent development platform. "We hope to make intelligent agent development more accessible through public testing and open sourcing," Zhou stated.

Guo Zhenxing, Vice President of Huawei China's Enterprise Business, assessed that 2025 was the first year of deep integration between industries and AI, with AI evolving from technological innovation to a core engine of industry transformation. He predicted that 2026 will be a year of deepening this integration, with AI penetrating core production scenarios, accelerating digital and intelligent upgrades, and unleashing new AI-driven productive forces.

The explosion of intelligent agent applications has led to a surge in computing power demand. Reports indicate that AI development is advancing rapidly, with daily token calls in the ToB sector exceeding 500 billion. This figure reflects both enterprises' active adoption of AI and their efforts to integrate technology into their operations, as well as growing concerns about computing power adequacy. According to a J.P. Morgan Securities research report, China's AI inference token consumption is projected to grow from approximately 10 quadrillion in 2025 to about 3,900 quadrillion by 2030, a more than 370-fold increase over five years.

The surge in token calls is directly driving growth in the domestic computing power industry chain. Against the backdrop of limited access to computing chips and the accelerated deployment of intelligent agents, ensuring sufficient computing power supply and reducing token costs have become core challenges for AI implementation. Hu Yuhai, President of Huawei Cloud Public Cloud, revealed that Huawei Cloud will launch a next-generation 3D data center this year. This technology breaks the traditional coupling architecture between cooling, power supply, and server racks, enabling distributed deployment across multiple floors. Additionally, Huawei Cloud introduced FlexNPU, a computing power operating system designed for flexible intelligence computing.

As AI agents increasingly integrate into core production systems, the "lobster farming" competition among tech giants is shifting from technical parameter comparisons to business value validation. Huawei Cloud announced a comprehensive upgrade of its partner policies, launching a dedicated 200 million yuan AI ecosystem fund. It also plans to collaborate with partners to establish industry-specific AI dream factories, with specialized zones for smart healthcare, embodied intelligence, smart manufacturing, and scientific computing to be rolled out this year.

In his speech, Zhou Yuefeng stressed that AI is developing at an unprecedented pace, with large models iterating almost weekly and AI's capability to handle long tasks transforming dramatically every four months. He noted that offline deployments make it difficult for enterprises to keep up with AI advancements, and that "public cloud is the optimal solution for AI implementation in enterprises" has become a consensus.

Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.

Most Discussed

  1. 1
     
     
     
     
  2. 2
     
     
     
     
  3. 3
     
     
     
     
  4. 4
     
     
     
     
  5. 5
     
     
     
     
  6. 6
     
     
     
     
  7. 7
     
     
     
     
  8. 8
     
     
     
     
  9. 9
     
     
     
     
  10. 10