DeepSeek has raised another round of funding. That, in itself, isn’t surprising—by 2026, fundraising by leading AI companies has become routine.
What is surprising is the contrasting stance taken by Alibaba and Tencent behind this very same round.
According to a report by 36Kr, one funding round reveals two radically divergent AI futures. Alibaba chose one path; Tencent, another. And increasingly, these paths no longer appear to be competing within the same race.
Alibaba’s Choice: All-in on Infrastructure
Alibaba’s AI strategy has always been clear: build infrastructure. From Qwen (Tongyi Qwen) to Alibaba Cloud, from model training platforms to compute services, Alibaba’s logic is simple: no matter who wins the AI application-layer race, they’ll all need to run on our infrastructure.
This mirrors Amazon AWS’s playbook. AWS doesn’t care whether Netflix or Spotify runs on its cloud—it focuses solely on delivering best-in-class cloud services.
In DeepSeek’s funding round, Alibaba played the role of an “ecosystem investor”: it invested not to gain control over any specific product, but to ensure the vitality of China’s broader AI ecosystem—and thereby drive demand for Alibaba Cloud’s compute resources.
This is a long-termist strategy. Its advantage? Risk diversification—Alibaba Cloud benefits regardless of which AI application ultimately succeeds. Its drawback? Slow returns—you won’t see explosive growth from any single product’s success.
Tencent’s Choice: Doubling Down on the Application Layer
Tencent’s strategy couldn’t be more different.
WeChat, QQ, games, video—Tencent’s core strength has never been technical infrastructure, but rather massive, real-world user scenarios. So Tencent’s AI strategy is: embed AI into every existing product, using AI to deepen its existing moats.
In DeepSeek’s funding round, Tencent acted more like a “strategic collaborator”—it hopes DeepSeek’s technology will directly power Tencent’s product portfolio.
This is a short-and-fast strategy. Its advantage? Rapid impact—AI can immediately enhance WeChat’s user experience, boost NPC intelligence in games, or improve video content recommendations. Its drawback? A clear ceiling—if the AI application-layer race ultimately pivots toward infrastructure dominance, Tencent risks falling behind.
The Core Divide Between the Two Paths
This isn’t merely about two companies choosing different strategies. It reflects fundamentally different judgments about the trajectory of the entire AI industry.
Alibaba is betting that AI will ultimately evolve into a utility-like infrastructure—akin to electricity, water, or coal—on which all enterprises build applications. So whoever controls the infrastructure controls the future.
Tencent is betting that AI’s value will manifest most decisively at the user end—whoever integrates AI most seamlessly into everyday life wins the market. So whoever controls user scenarios controls the future.
Both views are defensible. And precisely because both are reasonable, the divergence becomes especially intriguing—it signals that, at least for now, nobody knows what the final shape of the AI industry will be.
Implications for the Industry: No Need to Pick Sides—Just Observe
For AI entrepreneurs and investors, Alibaba and Tencent’s divergence is actually good news.
It means the AI industry is far from converging on a single development path. While tech giants continue debating whether infrastructure or applications matter more, startups gain wider windows of opportunity.
You can build a vertical AI application atop Alibaba Cloud. You can leverage Tencent’s user scenarios to launch an AI-native product. Or you can choose neither—and pioneer something entirely new, something the giants haven’t yet imagined.
The key is not to blindly follow whichever path appears more mainstream. Alibaba and Tencent’s divergence reminds us: the direction of the AI industry remains undecided—and it’s far too early to declare a winner.
On the surface, DeepSeek’s funding round is just one company raising capital. Beneath the surface, it’s a microcosm of China’s deepening strategic rift in AI.
And this rift may take several more years to resolve.