Do Son 
The debate over restricting open-source AI has flared up sharply. It now runs hot between Washington and Silicon Valley. In response, 25 technology firms and organisations have taken a public stand.
The group includes NVIDIA, Microsoft, and Meta. Together, they signed an open letter titled “Open Weights and American AI Leadership.” The message defends the future of open-source AI in forceful terms.
NVIDIA founder and chief executive Jensen Huang went further still. For this cause, he published his very first post on X. Microsoft chief executive Satya Nadella then shared the statement on his own social pages.
25 Signatories: Openness as the Foundation of US AI Leadership
The roster runs deep. Alongside Meta, Microsoft, and NVIDIA, it features IBM, Dell, Palantir, Mistral AI, Hugging Face, Perplexity, Mozilla, Y Combinator, and the venture firm Andreessen Horowitz.
The joint open letter makes a clear demand. It urges US policymakers not to impose premature limits on open-weight AI models. Such models can be downloaded, inspected, modified, and run on a user’s own hardware.
The letter presses several points. Open models widen participation in the AI economy. Moreover, they sharpen market competition and return control to users.
Huang’s Case for Open Models
Huang framed the stakes broadly in his post. He argued that AI will transform every industry, power every company, and be built by every nation. He added that open models strengthen security and cyber defence while accelerating innovation and adoption. In his words, the world needs both cutting-edge closed models and cutting-edge open ones, a view he laid out in his first-ever post on X.
Answering the Misuse Concern
Critics worry that open AI invites abuse. The signatories push back directly. Closed models guarantee no absolute safety, they argue, because closed systems still face hacking and misuse.
Worse, when a closed system fails, outsiders often cannot even detect the problem. Therefore, the firms contend that harnessing openness to reinforce security is the sounder response to risk.
Aimed at Trump Administration Policy?
The timing carries real weight. Recently, the Chinese startup Moonshot AI released Kimi K3, billed as the world’s strongest open model. That launch drew intense attention in Washington.
Senior figures in the Trump administration have hinted at possible controls on such competitive open models. US Treasury Secretary Scott Bessent voiced support for open-source innovation. However, he also warned that openness does not license free use of American intellectual property.
His remark suggested closer scrutiny of foreign firms. In particular, officials appear focused on model distillation and similar routes to US technology.
The letter therefore reads as a pointed message to government. It states plainly that US leadership rests not on any single frontier model, but on whether the country builds a strong, open ecosystem that reaches every industry.
OpenAI and Anthropic Sit It Out
One absence stands out. Neither OpenAI nor Anthropic, the pace-setters of frontier AI, appears on the list. Google and xAI likewise declined to sign.
The two camps have now taken shape. The open-source bloc, led by Meta and NVIDIA, wants ubiquity to expand its hardware and ecosystem reach.
The closed bloc, led by OpenAI and Anthropic, prefers a different path. It keeps its most powerful models, such as GPT-5.6 Sol and Fable 5, behind paid API walls. At times, it quietly backs tighter national-security controls on frontier AI.
Open Versus Closed Will Redraw the Tech Map
This letter is more than a public-relations gesture. It marks a fierce contest among rival interests along the AI supply chain.
Consider NVIDIA first. The more people download and deploy open models locally, the more GPU compute the market must buy.
Meta reads it differently. Free open models serve as its sharpest weapon against OpenAI‘s subscription grip, and they help Meta own the underlying logic of future social platforms.
Microsoft plays a subtler game. It remains OpenAI’s largest backer, yet it still signed the letter. That choice reflects hedged bets, including its own Phi open-source models, and a drive to protect Azure’s infrastructure edge.
The road ahead now hinges on Washington. How the government balances national-security controls against the rewards of open innovation will shape US-China AI competition. Ultimately, it may define the trajectory of the entire global AI industry for the next decade.
Support Our Threat Intelligence
If you find our CVE report and cybersecurity news helpful, consider supporting our work.
Buy Me a Coffee
PayPal
