Jensen Huang's Open Weights Gambit: A Playbook, Not a Plea
Jensen Huang didn't just stake NVIDIA's interests last week. He exposed a fundamental tension in how the AI industry defines security, sovereignty, and competition. An open letter, signed by tech titans including Meta, Microsoft, IBM, and Dell, argued that open models "strengthen safety and cybersecurity, accelerate innovation and diffusion, and enable sovereignty." But the real story here isn't about safety. It's about who controls the infrastructure. And who gets to profit from it.
Of course it was NVIDIA. They have the most to lose from a world where open-weights models—models anyone can download, modify, and run on their own infrastructure—get squeezed out by proprietary alternatives.
The Geopolitics of Open Weights
The letter arrives at a critical juncture. The US hasn't invested in open-weight models to the same extent as Chinese model houses. Meanwhile, DeepSeek, MiniMax, Z.AI, and Moonshot (Kimi) are producing models that are competitive with, if not superior to, their American counterparts. Thinking Machine Labs' nearly-billion-parameter Inkling model is the best open-weight model the US has to offer, and if benchmarks are to be believed, it's not even in the same ballpark as Kimi K3 — see our analysis of Kimi K3 and the Open-Weight Rebellion Against Silicon Valley's AI Monopoly.
This isn't just about market share. It's about national security. If the US restricts Chinese open-weight models, American enterprises have two choices: use mediocre domestic models, or pay premium prices for proprietary models from OpenAI, Anthropic, and Google. Either way, NVIDIA's addressable market shrinks.
And what good are all these "AI factories" that NVIDIA has helped prop up if there's no diverse ecosystem to fill them? The Trump administration apparently learned that lesson the hard way trying to get its hardware back into China. So instead, NVIDIA is lighting a fire under American model developers to get their act together.
NVIDIA's Leverage Play
Here's where it gets interesting. NVIDIA has announced billions of dollars in investments in AI startups, including $30 billion in OpenAI and $10 billion in Anthropic. These investments are often tied to infrastructure deployments—meaning NVIDIA benefits whether these companies succeed or fail. But if the US restricts Chinese open-weight models, these companies' options shrink dramatically.
The Wall Street Journal reported Sunday night that NVIDIA may make a far larger commitment to OpenAI: a $250 billion backstop to help the model-maker lease space from a planned $10 billion datacenter Softbank is building in Ohio. That's not just an investment. That's leverage.
The open letter may as well have been Huang's way of saying: "It would be an awful shame were something to happen to that cash, Mr. Altman. Maybe it's time to get back on the 'open' bandwagon that gave your company its name."
That's certainly what OpenAI, SpaceXAI, and Meta seem to have heard. On Friday, OpenAI CEO Sam Altman quoted Huang's X post and reiterated his support for open weights models. "I want the US to win in AI both in open source and proprietary models, and I am glad to see this," he wrote. Since the letter's release, both OpenAI and SpaceXAI were added as signatories.
The Open Source Backlash
Not everyone is on board. Anthropic CEO Dario Amodei argued that "Anthropic has never advocated for a ban on open-weights models"—just those produced by authoritarian governments. His primary concern? "The risk that authoritarian governments—not solely the Chinese Communist Party (CCP), although the CCP is clearly the most capable threat—build AI models that are more powerful than those built by the US, and use them to achieve permanent military superiority or perpetrate incredibly deep repression of their own people." For a deeper look at Anthropic's nuanced position, see The Open-Weight Paradox: Why Anthropic's Boss Wants Powerful AI — Just Not in Beijing's Hands.
But Amodei's secondary concern reveals a deeper anxiety: "Open-weights models — it does not matter whether they come from China or anywhere else — do potentially present a higher risk than closed models, because it is very difficult to apply guardrails to them or monitor their usage, and once weights are released they cannot be withdrawn."
So rather than banning open-weight models, Amodei is advocating for three measures: banning the export of high-end accelerators to China, cracking down on industrial-scale distillation operations (for which his company has previously accused Alibaba and other Chinese model devs), and mandating safety testing for sufficiently potent models.
It's not a ban on all Chinese models. It's a ban on those that didn't play by Anthropic's rules and are big enough to compete with his own. As of yet, Anthropic appears to be sticking to its guns and has no plans to enter the open weights model arena.
The Real Test: Investment, Not Rhetoric
These commitments may turn out to be little more than virtue signaling. We don't know yet. The real test isn't whether we see new open-weight models come out of SpaceXAI, Meta, or OpenAI—it's whether they invest enough so they can hold their own against Chinese models.
OpenAI is the most likely player to pull this off. OpenAI is certainly no stranger to open-weights models—GPT-2 fit the bill way back in 2020, and last year it emitted its first open model since then, GPT-OSS. At 20 and 120 billion parameters in size, the models were among the United States' most capable open models at the time, but they fell well short of OpenAI's proprietary models and were no match for the growing number of models coming out of China.
But now, GPT-OSS is practically geriatric—in the AI equivalent of dog years that is—celebrating its first birthday next month. If we're going to see a GPT-OSS-2, next month would be the time. Assuming they've actually been working on a successor. If they haven't, well, it's going to be a bit.
Meta's Chief AI Officer Alexandr Wang confirmed that Meta would be getting back into the open model race soon enough. At one point, Meta had gone all in on open-weight models as its key differentiator. But after Llama 4 failed to impress last year, the social network pivoted to proprietary models. Its first, Spark Muse, debuted earlier this year.
SpaceXAI CEO Elon Musk joined in on the Twitter quotefest: "This has my full support. Jensen is right." Like OpenAI and Meta, xAI, now part of SpaceX, is no stranger to open models. Earlier in the company's foray into LLMs, it committed to releasing older model weights to the public. Unfortunately, as is so often the case with Elon's pronouncements, his company hasn't followed through on that promise. Earlier this month, the conglomerate rolled out Grok 5 to the public, yet its most recent open-weight model on Hugging Face is still Grok 2.
The Cybersecurity Irony
Last week, the AI flag bearer admitted that its models powered an agentic cyberattack on Hugging Face's infrastructure. The popular model repo quickly launched its own AI-powered response, only to discover those same models' safeguards were getting in the way of the analysis. In a fitting bit of irony that went precisely against OpenAI's interests, Hugging Face was forced instead to rely on uncensored Chinese models to get the job done. The incident is explored in depth at When Frontier AI Escapes: The Hugging Face Hack and the Case for Open Models.
The timing here is worth noting: right about the time model devs were reportedly using Moonshot's Kimi K3 release to lobby White House officials into restricting Chinese model use, OpenAI's models were stirring up some drama of its own.
On Monday, NVIDIA unveiled the Open Secure AI Alliance, which aims to ensure defenders have the tools they need, including open frontier models and tools, to protect themselves from a new generation of cybersecurity threats.
In either case, what's good for NVIDIA's bottom line also happens to be in the best interests of enterprises regardless of whether they buy Huang's hardware or not. Open ecosystems and choice are never a bad thing.