Nvidia CEO Jensen Huang posted on X for the first time on July 24th, using the debut to share a letter Nvidia signed arguing that open AI models deserve as much support as closed, proprietary ones. The letter states that AI will transform every industry, power every company, and eventually be built by every country, and argues open models strengthen safety and cybersecurity, speed up innovation and its spread to more users, and let countries build AI systems under their own control rather than depending entirely on foreign providers. Huang's post frames the choice between open and closed models as false, saying the world needs both kinds of frontier systems rather than picking one over the other. The full letter is posted as a PDF on Nvidia's own site. Huang's move comes as Nvidia, whose GPUs power most AI training and inference worldwide, has increasingly positioned itself as a backer of open-weight models such as Meta's Llama family, even as it also profits from closed labs like OpenAI and Anthropic buying its chips. Taking a public stance now lands in the middle of an active policy fight in Washington and Beijing over whether open-source AI helps or threatens national security, a debate that could shape future export rules and government AI contracts.
Why it matters · Nvidia's own CEO is now on record backing open models even though the company's business depends on both open and closed labs buying its chips, a stance that could shape whether Washington tightens or loosens rules on open-source AI.
Worth asking · Is Jensen Huang's open-models stance a genuine policy position, or a hedge to keep selling chips to both open-source labs and closed frontier players no matter which side wins the argument?