# Nvidia, Palantir, Meta warn against ‘premature restrictions’ of open-weight models

> Source: <https://www.scmp.com/news/us/diplomacy/article/3361813/nvidia-palantir-meta-warn-against-premature-restrictions-open-weight-models?utm_source=rss_feed>
> Published: 2026-07-24 21:57:47+00:00

Advertisement

# Nvidia, Palantir, Meta warn against ‘premature restrictions’ of open-weight models

Top US tech companies urge policymakers in open letter not to undermine competition and drive more users to Chinese AI systems

3-MIN READ3-MIN

Listen

[Vincent Chow](/author/vincent-chow)in Washington

A group of American tech giants, including

[Nvidia](https://www.scmp.com/topics/nvidia?module=inline&pgtype=article),[Palantir](https://www.scmp.com/topics/palantir-technologies?module=inline&pgtype=article)and[Meta](https://www.scmp.com/topics/meta-platforms?module=inline&pgtype=article), signed a public letter on Friday calling for US leadership in open-weight AI models, an area that[China](https://www.scmp.com/topics/china?module=inline&pgtype=article)currently dominates.The challenge comes amid growing speculation that the Trump administration might introduce restrictions on open AI models.

These would aim to address national security risks of the fast-moving technology and combat the growing popularity of Chinese models, a move that critics say would instead hurt US innovation.

Rather than banning

[open-weight AI](https://www.scmp.com/tech/tech-war/article/3361142/why-chinas-open-weight-ai-model-kimi-k3-sparking-anxiety-silicon-valley?module=inline&pgtype=article), the US should build its own “strong, open ecosystem” and avoid “premature restrictions on open models that stifle competition or drive innovation overseas”, the three-page letter said.Open-weight models are AI models that are released for free on the internet, allowing anyone to download and customise them for different use cases.

Proponents say that open models promote innovation and adoption by lowering the costs of using

[AI](https://www.scmp.com/news/china-future-tech/ai?module=inline&pgtype=article)and allowing end users to run the models on their own servers, meaning that user data is retained locally.Advertisement

Select Voice

Select Speed

1.00x
