Português
EnglishIn 2026, during the World Artificial Intelligence Conference (WAIC), the “How to Harness Enterprise-Level Agents” special forum was held in Shanghai on July 19, hosted by DingTalk and located at the Shanghai Expo Center. This event focused on the implementation paths, organizational practices, and future development directions of enterprise-level agents, attracting hundreds of AI experts and leading industry representatives to explore the transformative landscape of enterprise productivity in the age of agents.
Attendees included Zhu Hong, head of DingTalk’s Core Platform Business Unit; Shri Narayanan, University Professor at the University of Southern California and member of the U.S. National Academy of Engineering; Gu Chao, Chief Security Officer of STO Express; Xu Jiaxin, General Manager of Kootai; Li Ren, Chairman of Zhongyi Co., Ltd. and founder of the Obena brand; Yi Zihan, Founder and Chairman of Songxianxian; Cong Zelong, Director of the Pisen AI Product Center; and Wei Jun, Chairman of Zhejiang Youkela Intelligent Technology, who shared cutting-edge insights and practical experiences on-site.
From Pilot Projects to Full Deployment: The Growth Trajectory of Enterprise-Level Agents
Over the past year, almost every company has been discussing, experimenting with, or even developing agents. However, how to truly integrate these intelligent entities into core business processes—transforming sporadic trials into stable, productive operations and scaling from single applications to full-fledged deployments—remains a major challenge for many enterprises.
Zhu Hong, head of DingTalk’s Core Platform Business Unit, pointed out that companies are no longer focused on whether they can build an agent, but rather on their ability to systematically generate, manage, and operate digital employees at scale. Since each agent incurs marginal costs, businesses must ensure they deliver tangible benefits. He anticipates that in the future, every competitive company will have numerous digital employees, requiring not just tools but an infrastructure that supports secure collaboration, continuous learning, and efficient operation of these agents.
Building a Collaborative Ecosystem: The Evolving Role of AI Dingling
In response to this trend, Dingling is redefining its position. Zhu Hong stated that over the past decade, Dingling primarily connected people and organizations; moving forward, its goal is to create a seamless network where humans collaborate with agents, agents interact with one another, and agents work alongside organizations, becoming the central platform supporting productive intelligent systems.
This is precisely the vision for AI Dingling over the next ten years—agent-friendly, developer-friendly, and data-asset-friendly. To achieve this, Dingling has launched an AI office intelligence platform, enabling every employee to create a personalized AI assistant, while accelerating the development of the AI Dingling ecosystem to help companies more effectively develop, integrate, and manage their digital workforce.
“Tools are just the starting point; true value lies in everyday practice—this is exactly why we organized this forum,” emphasized Zhu Hong. He expressed hope that, through the new-generation AI productivity platform, DingTalk can partner with more enterprises to jointly address the critical question of “how to effectively harness agents.”
Industry Examples: Deep Integration of Agents into Business Operations
During the enterprise AI practice-sharing session, representatives from logistics, retail, manufacturing, and smart technology sectors presented real-world cases of embedding enterprise-level agents into core processes.
As one of China’s largest low-cost express delivery providers, STO Express operates under a franchise model with approximately 350,000 employees, managing safety across both offline operations and online information systems. Chief Security Officer Gu Chao noted that AI is gradually shifting from an efficiency-enhancing tool to a key pillar of risk prevention.
For example, identifying safety hazards used to rely on manual photo reviews, often missing various potential risks. Today, with AI-trained skills and knowledge bases, the system automatically detects anomalies, suggests corrective actions, and promptly notifies responsible personnel via Dingling, tracking progress to form a complete “detection–resolution–closure” management loop.
Kootai is a rising retail brand specializing in whole-house storage solutions, characterized by high unit prices, demanding service standards, and a strong emphasis on customer experience, operating roughly 200 partner stores nationwide. General Manager Xu Jiaxin stressed that the true measure of AI’s success lies in its seamless integration into business workflows.
She cited content marketing as an example: previously, social media content heavily relied on individual creativity, resulting in inconsistent output. Now, AI handles trend analysis, copywriting, video editing, and livestream debriefing, while frontline teams use AI for automated customer responses and private-domain user outreach, significantly boosting content production capacity and consistency. Since deeply integrating AI into workflows in March of this year, within just four months, employees’ working styles have largely shifted to a dual-track collaboration model—AI on the left, tasks on the right.
Pisen was among the first companies to invest in AI hardware R&D, unveiling the world’s first desktop AI robot at CES this January. Director of the AI Product Center, Cong Zelong, explained that if engineers’ expertise remains confined to personal experience, innovation speed will be severely hampered.
Therefore, Pisen integrated its decade-long R&D system and proprietary component database into an AI platform. After inputting product requirements, the agent automatically extracts parameters, searches for suitable components, instantly calculates the BOM cost, proactively recommends alternatives when budgets are exceeded, and employs an adversarial review mechanism to identify gaps, ultimately generating preliminary hardware architecture diagrams and specification documents. This compresses what once took over a month into a week-long process. The team has also built a skill-sharing platform, transforming individual expertise into reusable digital assets, now boasting over 50 high-quality skills and exceeding 600 code submissions.
In addition, Li Ren, Chairman of Zhongyi Co., Ltd.; Yi Zihan, Founder and Chairman of Songxianxian; and Wei Jun, Chairman of Zhejiang Youkela Intelligent Technology, also shared real-world applications powered by Dingling AI. These cross-industry examples demonstrate the maturing journey of enterprise-level agents—from initial pilots to core business integration—highlighting Dingling AI’s profound impact on cost reduction, efficiency improvement, risk control enhancement, and organizational collaboration.
The Core of AI Implementation: Continuous Evolution of Organizational Capabilities
Concurrently with the conference, Dingling showcased its latest achievements in enterprise-agent and AI collaboration at Alibaba Group’s exhibition booth. Applications on display—including AI voice recorders, AI note-takers, AI spreadsheets, and the AI productivity platform—covered diverse scenarios such as meeting minutes, document processing, presentation creation, information retrieval, data collaboration, and workflow tracking, drawing large crowds eager to experience them firsthand.
The enthusiastic response at the booth underscored a common understanding: introducing AI into enterprises is not merely a technical issue, but involves reshaping organizational collaboration models, redesigning workflows, codifying knowledge and experience, and building long-term AI operational capabilities. The true value of agents lies not in flashy demonstrations, but in the subtle nuances of daily operations; not in isolated tools, but in the overall performance gains achieved across the entire organization.
DomTech is DingTalk’s official designated service provider in Macau, dedicated to providing DingTalk services to a wide range of customers. If you’d like to learn more about DingTalk platform applications, feel free to contact our online customer service or reach us by phone at +852 95970612 or email at [email protected]. With a skilled development and operations team and extensive market service experience, we’re ready to offer you professional DingTalk solutions and services!