In his first post on X, Nvidia CEO Jensen Huang spotlighted a joint effort by major technology firms to support open-weight AI models. The message stressed that such systems will reshape industries worldwide and foster broader innovation.

Huang noted that open models boost safety measures, speed up progress, and support national independence in technology. On Friday, Nvidia joined Microsoft, Meta, OpenAI, Mistral, and additional partners in releasing an open letter that asks American officials to avoid strict limits on these AI tools.

The statement comes as some officials in the current administration have signaled possible steps against Moonshot AI, the firm behind the leading open-weight system known as Kimi K3. Signatories argued that American success in AI should be measured by the strength of an open ecosystem rather than any single leading model.

Initially, OpenAI and Anthropic were not listed among the signers, though OpenAI was later added to the roster shared by Microsoft. The company offered no immediate comment on its participation. Unlike certain closed models from other labs, Kimi K3 is set to become freely downloadable and adaptable by users.

OpenAI leader Sam Altman expressed approval for the push to defend open-weight approaches and voiced a desire for the United States to lead in both open and closed AI systems. Elon Musk, though not a direct signer, voiced full backing for the position outlined by Huang.

The letter draws parallels to the rise of open software in the 1980s, which lowered barriers and built lasting American expertise. It points out that Chinese developers have taken a strong position in open-weight AI despite the United States long history with open-source tools. Mozilla and the Linux Foundation were among the organizations that added their names.

In the domestic market, Nvidia offers its Nemotron model as a notable open-weight option, while Meta continues work on an open version of its Muse Spark system through its Superintelligence Labs group. Other startups are also preparing releases in this space.

Recent comments from White House science policy director Michael Kratsios and Treasury Secretary Scott Bessent raised concerns over distillation practices, where less advanced models are trained on outputs from stronger ones. The letter acknowledges risks from improper extraction of value from closed systems but calls for targeted legal responses instead of broad bans on useful techniques.

By Friday evening, 32 entities had endorsed the letter, including the American Innovators Network, Andreessen Horowitz, Arcee AI, Arena, Black Forest Labs, Box, Cisco, Cohere, CrowdStrike, Dell Technologies, DoorDash, Emergence Capital, Fireworks AI, GitHub, Hugging Face, IBM, The Linux Foundation, Mariana Minerals, Meta, Microsoft, Mistral, Mozilla, Nvidia, OpenAI, Palantir, Palo Alto Networks, Perplexity, Reflection, Replit, ServiceNow, Telnyx, and Y Combinator.