41 companies, one message: open-weight AI is America's best bet

Forty-one organizations, including rivals OpenAI and Meta, made an unusual joint statement on July 24: open-weight AI models are not just a development choice, but a strategic asset for national security and economic competitiveness, echoing concerns that America's AI moat is draining. The open letter, titled "Open Weights and American AI Leadership," brings together cloud providers, chip makers, venture capital firms, and startups in a rare show of unity on a topic that usually divides them.

The Unexpected Alliance

The signatories list reads like a cross-section of the entire tech industry: AI21, AMD, Andreessen Horowitz, Black Forest Labs, Block, Cisco, Cloudflare, Cohere, CrowdStrike, Dell, DoorDash, Fireworks AI, GitHub, Google, Hugging Face, IBM, Meta, Microsoft, Mistral, Mozilla, NVIDIA, OpenAI, Palantir, Palo Alto Networks, Perplexity, Replit, ServiceNow, Y Combinator, and many more. The letter argues that open-weight models, which anyone can download, inspect, modify, and run on their own infrastructure, are the foundation for a thriving domestic AI ecosystem that diffuses into factories, hospitals, farms, and main street businesses. It frames openness as the path to broad adoption, rather than concentrating gains among a few frontier labs.

Benchmark Breakthrough: MiniMax M3

The coalition's argument arrives at a moment when the evidence for open-weight capability is stronger than ever. Chinese lab MiniMax released its M3 model, an open-weight system with a novel sparse attention architecture, a technique shown to deliver 7x faster decoding without quality loss, and native multimodal training that scored 83.5 on BrowseComp, edging past Opus 4.7. On coding benchmarks, M3 outperformed GPT-5.5 in real code tests, according to our coverage. Perhaps more telling was an autonomy test: M3 self-optimized a CUDA kernel from 7.6 percent to 71.3 percent peak utilization, and replicated an ICLR paper in 12 hours, all without human intervention. These results chip away at the assumption that open models must trail proprietary systems by a generation.

Economic and Strategic Rationale

The economic stakes are visible in the market reaction to other open-weight models. When Moonshot AI released Kimi K3, a 2.8 trillion parameter open-weight model, the combined estimated valuations of OpenAI and Anthropic lost $314 billion as investors confronted cheaper frontier AI from China. The coalition's letter argues that closing off open-weight development would hand the advantage to foreign competitors and slow the diffusion of AI into the broader economy. IBM's open-source agent framework and the $800 million Series C for an open-source AI startup underscore the industry's bet on openness as a scalable model, a trend visible in rising open model demand that overwhelmed Ollama's subscription model.

Cracks in the Consensus?

The very signatories that profit most from closed models signing this letter deserve scrutiny. OpenAI, which has built its business on proprietary systems (a strategy that contrasts with Claude Opus 5's approach of capping certain capabilities), may have a strategic and defensive incentive: if the regulatory pendulum swings toward restrictions, it could hinder the entire ecosystem, including its own downstream customers. Meta, whose Llama models are the most widely used open-weight family, has a direct stake in keeping the ecosystem open. Internal documents suggest Llama-5 will be a 1.2 trillion parameter dense model, further consolidating Meta's dominance in the open-weight space. For other signatories, the letter may serve as a hedge: let a thousand models bloom, and the best ones will float.

Policy Implications

The open letter gives policymakers a concrete document to cite, backed by the weight of 41 major organizations. As one executive from a signatory company put it off the record: "The fastest way to lose the AI race is to overregulate before we know what works. This letter says: let a thousand models bloom, and the best ones will float." Whether that argument persuades legislators who are fielding warnings about extinction risks and dual-use hazards remains an open question. But the coalition represents the industry's center of gravity, the part that believes open access is not just a development model but a national security play, a framing that resonates with Poland's state investment in AI as a policy signal. If the US adopts regulations that restrict open-weight distribution, it may find itself at odds with its own most competitive firms, and with the evidence that open models can already hold their own against the best proprietary systems.