On May 20, Kingdee held an AI summit in Shenzhen, released "AI-native products," and jointly released a set of "AI management" research results with Peking University HSBC Business School.
The choice of this timing is quite intriguing—the next day (May 21), the national-level AI agent evaluation standards seminar was submitted for deliberation. Kingdee spoke out a day earlier, seizing the initiative in the perceptual positioning of 'AI management'.
This matter itself is not news. What is noteworthy is: the AI narrative of domestic ERP vendors is shifting from "feature stacking" to "management restructuring". What does this shift mean for the competitive landscape of the ERP market?
Behind the 48.7 billion market, what are local manufacturers competing for?
In 2026, China's ERP market size is approximately RMB 48.7 billion, with local vendors holding a 68.4% share, up 12.2 percentage points from last year. Cloud-native penetration has exceeded 51.2%, and products deeply integrating AI account for more than 65%.
The numbers may look impressive, but on closer inspection, the core of competition has long since changed—
In the past, local vendors relied on cost-effectiveness + localization to compete with SAP and Oracle for market share. Now this logic is weakening, because SAP and Oracle are also doing two major things:
- SAP Sapphire 2026 unveiled the 'Autonomous Enterprise,' with more than 50 Joule agents directly executing business processes rather than merely assisting.
- Oracle is also pursuing a similar path, upgrading AI from an "assistant" to an "executor."
This means that if local manufacturers still stay on the narrative of "our AI can help you automatically generate reports," the gap will widen again.
So Kingdee's current strategy—upgrading the narrative from “features” to “management”—is the right direction. But whether it can actually be implemented still depends on a few things.
Kingdee's "AI Management": Concept Comes First, Real-World Implementation Still Needs Three Years to Prove Itself
According to public information, the core content of Kingdee's release this time consists of two parts:
- Product level: AI-native products (Kingdee AI Xinghan, Kingdee AI Xingkong, Kingdee Lingxi, etc.), emphasizing “AI-native” rather than “AI overlay”
- Research level: Jointly released “AI Management” research findings with Peking University HSBC Business School, elevating AI from a technical topic to a management topic
The brilliance of this strategy lies in: it shifts the dimension of competition from “whose model is stronger” to “who better understands the AI-driven transformation of enterprise management”. SAP speaks of “autonomous enterprise,” while Kingdee speaks of “AI management”—one leans toward a technological vision, the other toward a management methodology, with different audiences and discourse systems.
But there are several issues that need to be resolved—
First, what exactly does "AI-native product" mean? Does it mean generating code with AI from scratch, or natively integrating AI capabilities at the product feature level? Currently, there is no consensus in the industry on this concept. If Kingdee can provide a clear definition and comparable cases within this year, its first-mover advantage will be very significant.
Second, can the "AI management" research methodology be productized? If Peking University HSBC Business School's research results remain only at the white paper level, their impact will be limited; if they can be transformed into features in Kingdee products, such as management diagnostics, process optimization recommendations, organizational design recommendations, etc., the value will be much greater.
Third, how long is the time window? SAP's “autonomous enterprise” narrative has already spread globally, and Yonyou and Inspur are also promoting their respective AI products. Kingdee needs to deliver verifiable customer cases within 6-12 months, otherwise its mindshare advantage will be diluted.
What Are Yonyou and Inspur Doing
Expanding the perspective to the entire domestic ERP market, the differentiation among vendors in their AI narratives has gradually become apparent—
| Manufacturer | AI Core Narrative | Current stage |
|---|---|---|
| Kingdee | "AI Management" + AI-Native Products | The concept has been published, and cases are accumulating |
| Yonyou | Enterprise Agent Platform (YonBIP AI) | Platformization stage, emphasizing the ecosystem |
| Wave | Manufacturing AI + State-Owned Cloud Scenarios | Deeply cultivate vertical industries |
| SAP (China) | Self-operated enterprise + Joule Agent | Global narrative, adapted for localization |
Yonyou’s approach differs from Kingdee’s. Yonyou emphasizes an “enterprise agent platform”—essentially packaging AI capabilities into a platform that allows ISVs and customers to build their own agents on top of it. This approach leans more toward ecosystem logic, and represents a business model entirely different from Kingdee’s “I define the AI management methodology.”
On the other hand, Inspur has deep roots in manufacturing and state-owned enterprise scenarios, and its AI narrative is more concrete: it is not "what AI can do," but "to what extent AI can help you in steelmaking, assembly, and warehousing." This vertical approach is very effective in specific industries, but its ceiling is equally obvious.
Will national AI agent standards become a variable?
The National AI Agent Evaluation Standards Seminar on May 21 is another noteworthy event this week.
The standard was jointly issued by three departments: the Cyberspace Administration of China, the National Development and Reform Commission, and the Ministry of Industry and Information Technology. It covers four dimensions: value quantification, scenario lists, security boundaries, and a full lifecycle closed loop. Simply put, in the future, enterprises purchasing AI agents will have national standards to rely on.
For ERP vendors, this is both an opportunity and pressure—
The opportunity lies in: with the endorsement of national standards, the decision-making threshold for enterprises purchasing AI will be lowered. In the past, CIOs had to bear the risk of "whether this AI is actually reliable"; now that standards exist, they can use these standards to select and acceptance-test products.
The pressure lies in the fact that once the standard is implemented, it becomes an entry threshold. AI functions that do not meet the "value quantification" and "safety boundary" requirements in the standard may be excluded from government procurement and state-owned enterprise procurement.
Kingdee held a press conference the day before the standard was released, which was very likely a deliberate timing arrangement—first launching its own "AI Management" framework, striving to align its framework with the standard when the standard was officially released, and even to influence interpretations of the standard.
What is the actual impact on businesses?
Translating these dynamics into the everyday language of enterprise CIOs and IT leaders, they can roughly be boiled down to the following questions——
First, when selecting an ERP, the weight of AI capability assessment will continue to rise. IDC data shows that 44% of CIOs have already made AI capability a primary criterion for ERP procurement. This proportion will continue to increase, especially among state-owned enterprises and large private enterprises.
Second, the gap between “AI-native products” and “AI-added products” will become apparent within 18–24 months. The current difference may seem like mere marketing rhetoric, but the difference in underlying architecture will lead to later divergence in customization costs, iteration speed, and data integration capabilities. When choosing a product, it is advisable to ask the vendor explicitly: Was this AI feature added later, or was AI taken into account from the product design stage?
Third, after the national standard is implemented, the procurement of AI agents will have clearer acceptance criteria. CIOs should monitor the progress of this standard in advance and incorporate it into their AI procurement framework for the second half of 2026. Waiting until the standard is officially released before following up may delay procurement progress.
Ultimately, competition in the ERP space has never been about comparing feature lists; it's about "who understands my business better." AI makes this question even harder—because now we also have to answer "who understands how my business is being changed by AI."
