China's Vertical AI Push Is Coming for European B2B SaaS
Chinese giants are skipping the LLM arms race and going straight for your workflows.
What happened in China
China's tech giants are not competing to build the best foundation model. They are competing to own the workflow. According to IMD, Chinese companies are embedding AI directly into specific industries — e-commerce, travel, logistics, and healthcare — bypassing the horizontal layer entirely.
The signal is clear in the data. According to OpenRouter, a global AI platform, API calls to Chinese large language models surpassed U.S. levels for the first time as of March 2026. The share of Chinese open-source models in global API calls climbed from less than 2% at the end of 2024 to overtake U.S. models in total token usage. Cost advantage is cited as a primary driver.
Behind this sits deliberate state alignment. China's 15th Five-Year Plan (2026–2030) commits to doubling down on Digital China development — giving incumbents like Alibaba and Tencent a policy runway that most European companies cannot match.
How it works / the detail
The architecture matters. WeChat operates as a super app — messaging, voice, video, social, commerce — inside a single interface. Tencent and Alibaba are now integrating AI features directly into these killer apps, according to SCMP citing JPMorgan, which means AI is not a product add-on but a workflow layer baked into where hundreds of millions of users already spend their time.
Alibaba Cloud posted revenue of RMB 39.8 billion (approximately US$5.6 billion) in the quarter ended September 2025, up 34% year-over-year and above analyst estimates. Alibaba's cloud division drove overall company revenue 5% higher to 247.8 billion yuan in the same period, with a 16% rise in a related metric. That is the financial engine funding the vertical AI build-out.
The commercial logic is also distinct from European SaaS. According to VC Cafe, vertical AI competes for labor budgets rather than IT budgets — a fundamentally different and larger pool of spend than traditional SaaS captures.
European comparison / the gap
European B2B SaaS has largely built on horizontal platforms — CRM, ERP, project management — where AI is being bolted on rather than built in. Vertical AI is growing at a 23.9% CAGR while horizontal SaaS faces commoditization, according to BuildMVPFast. European startups are tracking this shift, but the depth of workflow integration being deployed by Chinese incumbents — with their existing super-app distribution and state-backed investment cycles — is structurally harder to replicate.
The cost gap compounds the problem. Chinese models' pricing advantages are a key driver of their surge in global API usage, according to multiple sources including 36Kr and OreateAI. European SaaS players building on top of expensive foundation models face margin pressure that Chinese-developed verticals, building on cheaper open-source infrastructure, do not.
What Chinese platforms have that most European startups do not: proprietary data at scale, embedded distribution through super-app ecosystems, and a Five-Year Plan that treats AI verticalization as national infrastructure.
What European founders/investors should do
Audit your workflow depth, not your feature list. The threat is not that Chinese models write better copy. It is that they sit inside the operational layer of an industry. If your product can be replaced by an AI agent embedded in a logistics or healthcare super-app, you have a moat problem, not a pricing problem.
Build proprietary data assets now. The window to accumulate differentiated, sector-specific training data in European markets — governed under GDPR and sector regulation — is a genuine structural advantage. Use it deliberately, not passively.
Watch the API usage curve. Chinese open-source models going from under 2% to majority global API share in roughly one year is not a trend — it is a signal of platform shift. Investors should track which European verticals are most exposed to cost-advantaged model substitution.
Target labor budgets, not IT budgets. If your vertical AI product is sold to a CIO, rethink the pitch. The Chinese playbook is to replace human workflow steps, not software line items.
Closing line
China is not building better SaaS. It is building the layer beneath it — and the clock on European differentiation is running faster than most founders think.
Sources
- IMD – China's 2026 Playbook: Redefining Global Tech Industry and Governance
- SCMP – China's Tech Giants Set to Lead AI Growth in 2026 Despite Chip Shortage
- Xinhua News Agency – Chinese AI Models Surpass U.S. in Global API Calls
- China's Official Government Website – 15th Five-Year Plan Digital China
- Bloomberg – Alibaba Revenue Exceeds Estimates Driven by AI and Cloud Growth
- OreateAI – The Great Token Shift: How Chinese AI Models Are Dominating the Global API Landscape
- VC Cafe – Vertical AI in 2026: The Good, the Bad, and the Ugly
- BuildMVPFast – Vertical AI Eating Horizontal SaaS in 2026
- 36Kr – Chinese AI Developments
- Yahoo Finance – Alibaba Revenue Tops Estimates
Sources
- China's AI Push Turns Tech Titans Into National Champions
- Why China's Tech Giants Are Racing | RoxomTV News - YouTube
- China's push for global AI dominance - YouTube
Sources
- China's AI Push Has Scale. But Substance? - Bloomberg.com
- China's AI strategy: How tech giants are leading the global race
- China AI industry looks unstoppable, but is it? | CNN Business
