Yang Zhilin, 34-year-old founder and CEO of Moonshot AI, is having a full-on internet main character moment on Weibo. Chinese users are calling him a “millennial genius,” and a widely shared hashtag, “Post-90s Tsinghua University genius crashed US stocks,” was the third-most-popular tag in Weibo’s tech hot searches as of 10 a.m. local time Wednesday, with 32 million views in 24 hours when Business Insider viewed it.

Here is the part that matters for decision-makers: Moonshot AI launched its latest AI model, Kimi K3, on Friday, and that release was followed by a sell-off of big US tech stocks like Google and Nvidia. Moonshot then claimed the launch drew so many new subscribers that it temporarily paused subscriptions to reduce load on its limited compute resources. In other words, this is not just online chatter. It is a demand-and-capacity story that is getting interpreted, in real time, as competitive pressure on US AI leaders.

To understand why the internet is treating this like a reckoning, you have to look at what Kimi K3 is. Moonshot AI’s Kimi K3 model is positioned to rival Anthropic and OpenAI models in coding, reasoning, and knowledge-based tasks. On top of that, Moonshot said it would release the model weights on July 27 so developers can fine-tune and build on it. In the AI market, that combination is a big deal: performance claims get attention, but “weights released” is what often accelerates ecosystem adoption, tooling, and downstream applications.

The Weibo obsession also reflects how fast China’s AI race has become a global scoreboard. A Weibo user from Beijing wrote under the tag on Tuesday that, “We once chased after Silicon Valley’s AI, but now a post-90s Tsinghua entrepreneur has unveiled a cutting-edge open-source model, directly shaking Wall Street.” Another user from Zhejiang called him a “top student from Tsinghua University,” adding that his “superior technology directly shook Silicon Valley’s AI monopoly.” The details being referenced may be more cultural than technical, but they point to the story’s fuel: a “homegrown” founder narrative tied to real market reactions.

There is also a capacity signal here that executives should not ignore. Moonshot AI said in a Sunday X post that the model launch drew so many new subscribers that it had to temporarily pause subscriptions to reduce load on limited compute resources. That is a classic startup squeeze, and it can cut two ways for a board: it can indicate early demand is outpacing supply, but it also shows operational bottlenecks. If your competitors can sell out their own compute and still attract millions of users, the competitive baseline changes quickly. If they cannot scale inference and infrastructure fast enough, that same demand can turn into churn or developer frustration. Either way, it becomes a strategic constraint, not just a PR headline.

International observers have taken note of what Kimi K3 is capable of, not only what it says it can do. Cloud platform Vercel’s CEO, Guillermo Rauch, said in a Friday X post, “This is the first time that an open model is ahead of all proprietary ones for this comprehensive web engineering benchmark.” SpaceX CEO Elon Musk said in a Saturday X post that Grok’s coming model “might be able to exceed Kimi,” thanks to its two-trillion-parameter training, and Kimi responded on Weibo, “Welcome to the 2 Trillion+ Club.” For founders and investors, that matters because it frames the competitive narrative as “open vs proprietary,” and then as “benchmarks vs benchmarks,” with the internet acting like an amplifier.

The market interpretation is also influenced by what happened last year. The article notes that the Kimi K3 release followed a sell-off of major US tech stocks reminiscent of the shake-up caused by China’s DeepSeek AI startup last year. In other words, investors are not starting from a blank page. They have a recent memory of how Chinese model launches can trigger rapid sentiment shifts toward US tech incumbents, whether those shifts reflect long-term fundamentals or short-term uncertainty. Either way, that volatility can affect capital allocation decisions, procurement priorities, and hiring plans across the AI supply chain.

Meanwhile, Moonshot is leaning into the “open” playbook while it manages scarcity. The July 27 release of weights to allow fine-tuning and building could widen adoption beyond end-user chat, pulling developers into the model’s orbit. For boards of AI companies, that is the second-order effect that often gets underestimated: the real competitive moat might not be just the base model. It can be the developer ecosystem that forms around it once weights are available, especially if benchmarks and perceived capability are strong.

So what should executives take from all of this? In a short span, Yang Zhilin’s Kimi K3 launch combined viral momentum, open-model distribution, and a real operational bottleneck significant enough that Moonshot paused new subscriptions. Then it coincided with sell-offs in US tech stocks like Google and Nvidia. Whether you view it as temporary market overreaction or the early signal of a lasting shift, it is hard to deny the direction of travel: the next phase of the AI race is being written in real time, by model launches that spread through both developer networks and mainstream market headlines.