Moonshot AI’s July 17 release of Kimi K3 has confronted Washington with an awkward fact: a Chinese open-weight model can approach leading American systems in coding and agentic tasks while charging less, though we challenge the simplicity of that premise here. On July 20, the White House was considering options to block or sharply restrict American companies’ use of Chinese models.
The primary concern is national security rather than profit. Economically, however, procurement bans, Entity List measures, or onerous hosting rules could behave like tariffs. Washington already uses trade barriers to keep Chinese EVs from undercutting domestic manufacturers. Restricting access to Chinese open-weights would produce a similar effect; it would remove lower-cost competition and preserve the price umbrella beneath proprietary American models.
Pixabay/Public Domain
The public-market exposure follows ownership and distribution. Amazon (NASDAQ:AMZN) may invest up to $25 billion in Anthropic and distributes Claude through Bedrock. Alphabet (NASDAQ:GOOGL) owns Gemini and reportedly plans to invest up to $40 billion in Anthropic. Alibaba (NYSE:BABA) develops Qwen and backs Kimi’s parent, Moonshot. Restrictions would not make Alibaba (NYSE:BABA) an immediate earnings casualty, but they would surely narrow the U.S. market for both Chinese model families.
Microsoft (NASDAQ:MSFT) has the clearest exposure to the frontier-intelligence theme. The company’s latest quarterly filing placed its OpenAI interest at around 27% on an as-converted basis. OpenAI’s March 31 funding round valued the company at $852 billion post-money, implying a stake worth roughly $230 billion, although Microsoft has not reported that amount as the investment’s carrying value.
OpenAI has committed to purchase another $250 billion of Azure services, while Microsoft retains revenue-sharing rights through 2030 and a non-exclusive IP license through 2032. Kimi’s coding strength also challenges GitHub Copilot more directly than it challenges most Big Tech products. Meanwhile, Microsoft’s expanded Mistral partnership gives Azure a Western open-weight alternative if Chinese models are excluded.
Hedge-fund interest is pretty broad but cooling. Among funds tracked by Insider Monkey, 282 reported Microsoft positions at the end of the first quarter, down from 312 three months earlier. Bill Ackman’s Pershing Square nevertheless opened a new position after the stock’s decline. Short interest stood at only 1.2% of the float on June 30, suggesting little outright bearish conviction.
We do want to mention, though, that policing open-weights models is not without challenge. Once open weights are published, they can be copied, modified, and served from infrastructure outside the United States. Washington could police federal contractors and domestic cloud providers, but preventing the files themselves from circulating would be considerably harder.
While we acknowledge the risk and potential of MSFT as an investment, our conviction lies in the belief that some other AI stocks hold greater promise for delivering higher returns and doing so within a shorter time frame. If you are looking for an AI stock that is more promising than MSFT and that has 10,000% upside potential, check out our report about the cheapest AI stock.
READ NEXT: 33 Stocks That Should Double in 3 Years and Cathie Wood 2026 Portfolio: 10 Best Stocks to Buy.
Disclosure: None. Follow Insider Monkey on Google News.
