Nvidia, Microsoft, Meta, Palantir, and more than 10 other companies sent a letter on Friday to policymakers, advocating against "premature restrictions" on open-weight artificial intelligence models. They argue that such regulations would "stifle competition or drive innovation overseas." This move comes amidst a heated debate regarding open-weight AI, particularly intensified by concerns raised earlier this year by the Trump administration.
The U.S. Treasury, through an official, expressed on Tuesday that the Trump administration is investigating whether Chinese AI companies, such as Moonshot AI, have engaged in "industrial-scale distillation attacks" to develop their models, like the Kimi K3, potentially leveraging stolen technology. Policy Director Michael Kratsios specifically accused Moonshot AI of distilling Anthropic's technology and possibly using Nvidia chips for training. These concerns have led to discussions about potential export violations for Nvidia's Blackwell chips.
However, the tech companies, including Nvidia CEO Jensen Huang and Microsoft CEO Satya Nadella who shared the letter, cautioned against rash actions. They asserted that open-weight models enhance competition and ensure that AI benefits are "broadly shared rather than concentrated in a few hands." The letter also stated that relying solely on closed models is not inherently safe, as they can be misused or breached, and concentrating advanced AI capabilities in a few closed models compounds these risks.
The letter explicitly calls for addressing concerns about unlawful distillation through "targeted legal and commercial frameworks" rather than "sweeping restrictions on techniques that play an important role in AI innovation." The companies emphasized that U.S. AI leadership will be judged by its ability to build a strong, open ecosystem that permeates every sector, crucial for national innovation and prosperity. Notably, OpenAI and Anthropic, both valued for massive potential IPOs this year, did not sign the letter.
