Can It Win the Enterprise AI Agents Race?

• Alibaba’s Wukong platform embeds enterprise AI agents directly into business workflows via DingTalk
• Wukong enters a crowded field where Tencent and ByteDance are already racing to own the enterprise AI agent stack

Alibaba has unveiled Wukong, an AI-native platform designed to coordinate multiple enterprise AI agents across business workflows. The pitch sounds compelling, but the platform is currently available only to invited beta testers, which means the gap between what Alibaba is promising and what enterprises can actually deploy remains wide open.

Wukong is available as a standalone desktop application or embedded within DingTalk, Alibaba’s workplace collaboration platform. DingTalk already serves more than 26 million corporate users, giving Wukong a ready distribution channel that most new enterprise tools would spend years trying to build.

The platform will progressively connect with Slack, Microsoft Teams, and WeChat–a cross-platform ambition that matters for APAC enterprises running mixed-tool environments. The product is designed around multi-agent coordination: rather than a single assistant responding to prompts, Wukong assigns interconnected tasks across agents operating within a single interface.

Alibaba’s broader ecosystem–Taobao, Tmall, 1688, Alipay, and Alibaba Cloud–will be progressively integrated as modular agent skills, alongside third-party integrations. The platform also ships with One-Person Team (OPT) solutions spanning ten verticals, including e-commerce, manufacturing, legal services, finance, software development, and others.

The ambition is notable, but ten industry verticals at the beta stage present a significant scope for validation, and Alibaba has not disclosed how many enterprises have tested these solutions or what results they’ve seen.

Enterprise AI agents enter a crowded field

What the press release glosses over is that Wukong arrives in a market that is already moving fast. DingTalk itself launched Agent OS back in December 2025–an operating system designed specifically for AI agents, alongside enterprise hardware called DingTalk Real, intended to give agents a physical execution environment in complex enterprise settings.

Wukong, developed by the team behind DingTalk and now operating under the newly established Alibaba Token Hub (ATH) Business Group, is the next layer on top of that infrastructure. The competition isn’t sitting still either.

Tencent recently launched WorkBuddy, a desktop AI agent for workplace automation with OpenClaw compatibility and over 20 skill packages covering invoice processing, report generation, and data tasks. Tencent, Alibaba, and ByteDance’s Volcano Engine have all integrated OpenClaw into enterprise products, with the open-source framework fast becoming a standard compatibility layer for AI agents in China.

China’s AI agent sector is projected to scale from under US$1 billion in 2024 to over US$30 billion by 2028, and the enterprise segment is where the structural money is heading. Gartner estimates that 40% of enterprise applications will adopt task-specific AI agents by 2026, up from less than 5% in 2025.

The enterprise distinction that actually matters

The more interesting question isn’t whether Alibaba is late; it isn’t, given its own prior moves, but whether Wukong is solving a meaningfully different problem than the consumer agentic wave. Platforms like Qwen and Doubao have been racing toward transaction completion and device-level control.

Alibaba’s Qwen already allows users to complete purchases across Taobao and Fliggy and process payments through Alipay, all within the chatbot interface. Xiaomi’s micLaw, meanwhile, is embedded at the operating system level, capable of reading messages, controlling smart home devices, and coordinating across Xiaomi’s device ecosystem.

Wukong’s pitch is that enterprise AI agents belong inside the workflows where business decisions actually happen, within the security boundaries that corporate IT teams require. That’s a reasonable argument–and the DingTalk integration gives it more structural grounding than a cold launch would have.

Whether that focus holds up once it scales beyond beta is the question Alibaba hasn’t answered yet. The DingTalk distribution advantage is real, but so is the competition and in a market where Tencent and ByteDance are moving just as fast, a tightly scoped enterprise pitch only stays differentiated until someone else makes the same one.

Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. This comprehensive event is part of TechEx and is co-located with other leading technology events, click here for more information.