Satya Nadella says trust in US technology will outweigh the allure of cheap Chinese models

Microsoft CEO Satya Nadella says trust in US AI technology and its robust training ecosystem will keep it competitive against Chinese products.

Nadella was speaking to CNN’s Fareed Zakaria on an episode of “GPS on CNN” that aired on Sunday.

Zakaria asked Nadella whether Moonshot AI’s Kimi K3 — a model that upset US tech stocks — and Chinese technologies have an edge over American products, because they’re “much cheaper” and can do what most companies need.

Nadella responded that the US has taken an “ecosystem approach” to shore up the competitiveness of its digital sector. “It’s been about trust in our technology, right?

So, for example, if you take the Chinese models, guess where these models run?

They run on a lot of the hyperscalers — that are American — all over the world,” Nadella said.

He added that because the Chinese models are open weight, US companies can monitor, test, and post-train them. “I think that one of the things that’s lost in this is the understanding even of how the US became competitive in the first place,” Nadella said. “We became competitive because we took our technology, we built a rich ecosystem where others could participate, in other countries, and trust our stewardship of this ecosystem,” he added. “And as long as that remains, we will absolutely be competitive, and we will win.” Microsoft has big stakes in the US AI game.

The tech giant holds about 27% of OpenAI’s for-profit public benefit corporation.

American AI companies face fierce competition from Chinese rivals, some of which have released bombshell products that roiled tech stocks.

Last year, the emergence of the Chinese AI lab DeepSeek prompted a $1 trillion tech sell-off, one of the first major signs that China could rival US AI capabilities at a much lower cost.

In July, Moonshot AI’s new AI model, Kimi K3, made waves in the tech sphere for its coding, reasoning, and knowledge work abilities, all of which could rival those of models from Anthropic and OpenAI.