In His First Twitter Post, NVIDIA CEO Shows Big Companies Pushing to Keep AI Models Publicly Downloadable

What are open-weight AI models, what is NVIDIA's CEO first X/Twitter post, and what does China have to do with all this? These and other questions will be answered in this article, as the world of AI has been busy these past several days.

Jensen Huang, founder and CEO of NVIDIA, joined X/Twitter in June and has just shared his first post, sharing a letter on "why open models matter," signed by some of the biggest tech companies in the world: Google, AMD, Cisco, Cloudflare, GitHub, Microsoft, Meta, IBM, Dell Technologies, Palantir, Hugging Face, Mistral, Andreessen Horowitz, Y Combinator, Perplexity, ServiceNow, CrowdStrike, and the Linux Foundation.

In the document, called Open Weights and American AI Leadership, he urged President Donald Trump to back open-weight AI models – models that anyone can download, inspect, modify, and run on their own machine.

He says that open weights fuel competition and let companies build on advanced models without training from scratch. He argues that they give customers control over their data and make sure that users don't become "locked into a single provider."

In terms of security, NVIDIA admits that when it leaves the creator's hands, an open-weight model is impossible to control, but "the right response to this risk is not to prohibit open weights. In a world where cybersecurity attackers use advanced AI, defenders need access to models with comparable capabilities so they can detect, simulate, and respond to emerging threats. Open models broaden defensive capability, increase transparency, and allow vulnerabilities to be discovered and remediated across many teams."

In conclusion, the company warned that "policymakers should be careful not to conflate legitimate model-development techniques with misappropriation" – one of the biggest issues of generative AI. Distillation – the practice of using one model's outputs to help train another – is a great form of improvement, while trying to extract value from closed models raises "legitimate concerns," which "should be addressed through targeted legal and commercial frameworks rather than sweeping restrictions on techniques that play an important role in AI innovation."

The US, like many other countries, is working on ways to integrate AI into the law, and this letter is meant to push the government in the right direction, at least according to those who signed it.

There might be one obvious reason NVIDIA asked for support: days before the letter was written, Chinese company Moonshot AI reportedly gained access to NVIDIA GB300 systems to build its latest AI model, Kimi K3, despite export bans.

This is the largest open-source model, with 2.8 trillion parameters, and it surpassed Anthropic's Fable 5 and OpenAI's GPT 5.6 Sol on some parameters. NVIDIA's most advanced AI chips are not allowed to be exported to China, and this violation is why the tech giant emphasized the benefits the US will get by releasing pro-open-weight models.

Who do you think will win this AI battle and why does it feel like a Fallout game? Don't forget to subscribe to our Newsletter, join our 80 Level Talent platform and Discord, and follow us on Twitter, LinkedIn, Telegram, and Instagram, where we share breakdowns, the latest news, awesome artworks, and more.

Are you a fan of what we do here at 80 Level? Then make sure to set us as a Preferred Source on Google to see more of our content in your feed.