Global Power Plays

Chinese AI models push into the U.S. by being cheaper and ‘good enough’

Chinese AI models are growing in the U.S. because they cost less and work well for most users.

Why this matters: San Francisco-based Raffi Krikorian, the chief technology officer at Mozilla, which runs the Firefox browser, switched to Chinese AI startup Moonshot’s Kimi K3 for many of his day-to-day activities within days of the new, powerful model’s launch more than a week ago.

What happened

New Chinese AI models are getting more use in the United States. People and some companies are choosing them because they cost a lot less than U.S. models. That includes developers, small businesses, and a few big tech users who need cheaper tools.

The U.S. government and some AI firms say parts of those models might copy U.S. systems. Beijing denies it. The claim has led to warnings about more export limits and public complaints from U.S. companies.

Who wins here

Chinese AI startups win by selling lower-cost models. They grab users who care about price over tiny gains in accuracy. Independent software makers also win: cheaper models cut their running bills a lot.

Big U.S. AI firms lose some customers and feel a squeeze on price and talent. U.S. policymakers gain leverage to tighten rules or push new controls if they choose to act.

How the play works

works on price and access. Chinese teams deliver models that run for pennies per million tokens. That is far cheaper than many U.S. options. Lower cost makes large-scale features, like running automated agents, practical for small buyers.

Another mechanism is openness. Some Chinese models are easier to run or modify. That lets developers plug them into apps fast. When tools are cheap and easy, they spread quickly.

Why it matters

Public costs are clear. If critical apps run on cheaper foreign models, the U.S. loses control over a chunk of the AI stack. That raises questions about safety checks, data rules, and who answers if something breaks.

There’s also an economic cost. U.S. firms may see lower margins and slower hiring. And policymakers face a choice: curb imports and raise costs, or let cheaper tech stay and risk losing influence.

What to watch next

Watch for U.S. export moves and more public claims of copying. New tariffs or chip limits would raise prices fast. Also watch adoption in developer tools and crypto firms, where cost matters most.

Track usage data from public model registries and cloud platforms. Big shifts will show up there first, and they will shape the next policy fights.

LensGlobal Power Plays
TypeReporting
PublishedJuly 26, 2026
Read time3 min read
SourceIndependent
Where the facts come from

The facts in this story were first reported by Independent. What you're reading here is our take on what it means for power and for you.

Read the original at Independent
Related topics

More stories on these topics

news analysisglobalmedia
Subscribe for moreExplore this lensBrowse all issues