Washington has repeatedly framed artificial intelligence as a strategic contest with China. Yet one federal search system appeared to include a China-built model.
That contrast drew attention after an image circulated online appearing to show the Federal Register's search interface offering Alibaba Cloud's Qwen as a selectable option.
Here's what to know
According to Futurism, scrutiny of the setup began when a Switzerland-based commodity portfolio manager posted a screenshot of the Federal Register search page.
US government officially distilling Chinese open source AI models 😉 pic.twitter.com/kQ9y6uPy25
— Yet another commodity guy (@tleilax___) September 15, 2026
The image showed the tool set to a default "semantic" mode and listed two entries connected to "Qwen3:0.6B." One of those entries was labeled a "hybrid" model and carried the description "512D, Recursive Splitting, Distilled, Prefixed."
Futurism said it later removed the Qwen-related options. A September 16 archived version of the page, however, still showed both search modes.
Qwen was created by Alibaba Cloud, and the outlet noted the episode surfaced amid escalating talk of an AI race between the U.S. and China.
TCD Picks » Quince Spotlight
💡These best-sellers from Quince deliver affordable, sustainable luxury for all
More background
Chinese companies' open-source models are increasingly being treated as lower-cost alternatives to pricier systems from major U.S. AI developers.
"Even the United States government is unable to afford the cost per token charged by American AI companies," an X user assessed.
As Futurism noted, the Federal Register contains public information, so the reported presence of Qwen does not, by itself, suggest that sensitive data was siphoned elsewhere.
Still, the fact that a government entity felt the need to offer up a Chinese AI model shows both that it is more economical and capable of the task. That doesn't necessarily align with rhetoric from government officials or domestic tech companies who insist the U.S. is ahead.
"I thought we were winning??" a user asked.
What's being done?
The Federal Register example may have been temporary, as it was taken down quickly after the situation went public.
Whether that was because the Chinese model could use search history for malicious purposes, or simply because it didn't look good, is unknown.
As Futurism noted, many models built by domestic companies are contracted by the government at very high prices. This episode showed that some government entities might simply be looking for a better deal for their LLM needs.
In that way, the outlet linked this incident to a wider trend of U.S. companies choosing cheaper Chinese models over more expensive domestic ones.
Where can I learn more?
Even as the two countries compete for supremacy, there's a widespread push to tap the brakes on AI in both America and China.
• In Washington and Beijing, safety advocates pitched prosperity while urging a pause on frontier AI.
• In China, a court ruled it illegal to replace workers with AI purely to cut costs.
• Public backlash intensified as AI is coming for jobs, water, electricity, and security.
Get TCD's free newsletters for easy tips, smart advice, and a chance to earn $5,000 toward home upgrades. To see more stories like this one, change your Google preferences here.
