Even as China’s broader economy slows, stock market funds continue to pour into the AI supply chain, including artificial intelligence, semiconductors, software and cloud computing.

According to Cryptopolitan on May 24, China’s April retail sales growth hit its lowest level since reopening after the COVID-19 pandemic, but stock market interest focused on AI and hard tech rather than domestic consumption plays.

At the center of the current investment flow are companies that align with Beijing’s push for technological self-reliance. That means semiconductors, high-performance components, AI models and software firms are trading more strongly than mall, dining and hospital-related stocks. Richian Ren of WisdomTree saw the tech growth narrative continuing. He noted that while many companies in the AI ecosystem are generating good profits, their scale alone is not enough to lift China’s overall economy. He called the current situation "very, very uneven."

The trend also shows up in the return gap between mainland and Hong Kong markets. The CSI300 index of large-cap stocks in Shanghai and Shenzhen is up nearly 5 percent this year, while Hong Kong's Hang Seng Index is little changed. The fact that many Chinese AI-related hardware makers are listed on the mainland A-share market rather than in Hong Kong is also affecting the flow of funds.

ByteDance and Huawei, major unlisted private companies in China, are hard for stock investors to access directly. Instead, more options have emerged as Chinese chip makers, AI model developers and advanced parts firms have entered the stock market one after another. Leonid Mironov's fund holds Tencent Holdings and Alibaba Group as key positions, and also includes hardware companies such as Shanghai-listed Anji Microelectronics.

Leonid said the market has not fully priced in the impact of policy support on the profitability of small and mid-cap tech stocks. "People don't properly see how fundamentally policy has helped the profit and loss of these small and mid-cap names," he said. He is not, however, betting indiscriminately across AI model companies. He said he would approach Zhipu and MiniMax after more clearly confirming customer retention and the sustainability of their business models.

By contrast, Morgan Stanley gave overweight calls on Zhipu AI, MiniMax and Alibaba. It also gave an overweight call on Cambricon Technologies and set a target price of 2,000 yuan. That signals increasingly selective investment in China’s AI theme across hardware and model companies.

Price competition is another pillar supporting China’s AI investment narrative. Hangzhou startup DeepSeek maintained a 75 percent discounted price applied to its V4 Pro for one month after launching the V4 series. The V4 series consists of the V4 Pro and the lightweight V4 Flash. DeepSeek's strategy is being read as a move to raise its profile not only in performance competition but also in cost competition.

Third-party benchmarking firm Artificial Analysis rated V4 Pro as one of the world’s top models by intelligence per dollar. The assessment reflects not only raw performance but also how much output users can get for the same cost. It also highlighted that price efficiency is an important variable when large-scale AI model operating costs are high and computing resources are limited.

DeepSeek's official API prices start at $0.0036 per 1 million cached input tokens and $0.87 per 1 million output tokens. According to Artificial Analysis, it costs about $268 to run the Intelligence Index benchmark with the model. Doing the same work with OpenAI's GPT-5.5 and Anthropic's Claude Opus 4.7 would cost 12 times and 19 times more, respectively.

This cost structure is a direct factor not only for software developers but also for exchange operators, trading firms and AI tool developers. The burden of output costs can rise as token usage accumulates. DeepSeek is not the only Chinese company to make the list in the cost-efficiency race. MiniMax’s M2.7 and Xiaomi’s Mimo V2.5 Pro were also included.

China’s AI investment boom is leaning more on technological self-reliance, cost competitiveness and potential policy benefits than on expectations for an economic recovery. The market focus is therefore likely to turn to whether individual companies can actually retain customers and sustain profitability, and how long low-price strategies can work in the global AI competition.