China has proven it can build world-class artificial intelligence models. What it hasn't proven is that the companies behind those models can turn technical brilliance into lasting profits. For investors, that distinction is everything.
In July, we argued that the model itself isn't the moat—the durable competitive advantage that protects a business. Last month, we applied that test to China's AI sector, and the industry has given us more evidence to consider. Here are three things that are still worth watching.
1. The Gap Between Technical and Commercial Leadership
By every visible measure, China has caught up technologically. Moonshot's Kimi K3, for example, now sits just three points behind leading U.S. models on certain benchmarks—a narrow margin that would have seemed unthinkable a year ago. Chinese labs like Moonshot, Baidu, Alibaba, and others are producing models that rival the best in the world.
But being the smartest model in the room doesn't automatically mean being the most profitable company. In the U.S., even the leaders in AI—companies like OpenAI, Anthropic, and Google—are still wrestling with how to monetize their models. The costs of training and running these systems are enormous, and competition is fierce. China's AI companies face the same challenge, often with even thinner margins and a more price-sensitive market.
The key question for investors isn't whether Chinese models are good. It's whether the companies making them can capture enough of the value they create. So far, the evidence is mixed. Many Chinese AI firms are still burning cash, and the path to profitability is unclear.
2. Who Actually Gets Paid in the AI Value Chain?
When the smartest model changes frequently, the model itself becomes a commodity. That's a problem for model-makers, but it can be an opportunity for companies that use AI to improve their existing products or services.
In China, that might mean software companies that embed AI into their enterprise tools, or consumer platforms that use AI to boost engagement and advertising revenue. It could also mean the companies that provide the infrastructure—chips, servers, and data centers—that AI depends on. These players may be better positioned to profit than the model-makers themselves.
Investors should also keep an eye on how Chinese AI companies are trying to differentiate. Some are focusing on vertical applications, like healthcare or finance, where specialized models can command higher prices. Others are building ecosystems that lock in users and make switching costs high. These strategies are more likely to create durable value than simply having the best benchmark score.
3. The Regulatory and Geopolitical Overhang
Chinese AI companies operate in a unique environment. The government has been supportive of AI development, but it also imposes strict rules on data, content, and security. These regulations can limit what companies can do and add compliance costs.
Geopolitics is another factor. U.S. export controls on advanced chips have made it harder for Chinese companies to access the most powerful hardware. While Chinese firms have responded by developing their own chips and optimizing their software, the constraints are real and could slow progress.
For investors, these risks mean that Chinese AI stocks can be more volatile than their U.S. counterparts. A single policy shift or trade restriction can move the market. That's why it's important to understand the broader context, not just the technology.
What It Means for Investors
The takeaway for everyday investors is straightforward: don't assume that the best AI model translates into the best investment. In China, as elsewhere, the companies that win are likely to be those that can monetize AI effectively, build defensible positions, and navigate the regulatory landscape.
That doesn't mean there are no opportunities. But it does mean that investors should look beyond the hype and focus on fundamentals. Watch for signs of revenue growth, margin improvement, and clear monetization strategies. Also, keep an eye on the broader market conditions—like the recent tech slide in China—which can affect even the most promising companies.
AI is a transformative technology, but it's still early days. The winners will be those who can build sustainable businesses, not just impressive models. As always, diversification and a long-term perspective are your best allies.
