Summary: Twenty-four technology companies and organizations, including Meta, Microsoft, NVIDIA, IBM, Dell, Palantir and Y Combinator, have signed an open letter urging U.S. policymakers to protect open-weight AI models from restrictive regulation. The coalition argues that open models lower costs, encourage competition, reduce vendor lock-in and support innovation across the AI industry. The letter also defends model distillation as a legitimate machine learning technique and arrives ahead of expected federal AI policy discussions.
It's rare to see Meta and Microsoft publicly aligned on an AI policy issue.
Even more unusual is seeing them joined by companies and organizations as diverse as Palantir, Mozilla, Hugging Face, Y Combinator and the Linux Foundation.
Yet this week, 24 organizations signed an open letter urging Washington not to impose restrictive regulations on open-weight AI models, arguing that doing so would reduce competition and slow innovation across the industry.
The coalition represents nearly every layer of the AI ecosystem, from chip manufacturers such as NVIDIA and Dell to model developers including Hugging Face and Mistral, alongside startup investors Andreessen Horowitz and Y Combinator.
Their shared message is simple: keep advanced AI accessible.
Open AI Models vs Closed Systems
At the center of the debate is whether advanced AI models should remain openly available or be controlled through proprietary platforms.
Open-weight models publish their trained parameters, allowing developers to download, modify and run them on their own infrastructure.
Closed models, such as those offered through commercial APIs, remain under the control of their developers and are typically accessed through paid services.
The coalition compares today's debate to the open-source software movement of the 1980s and 1990s, arguing that open technologies historically encouraged greater innovation, competition and lower costs.
Why the Coalition Supports Open Models
The letter makes three primary arguments.
First, open-weight models reduce the cost of AI adoption by allowing startups, universities and public institutions to build applications without training frontier models from scratch or relying entirely on commercial APIs.
Second, open models increase competition across the AI ecosystem by preventing vendor lock-in and allowing organizations to deploy AI on infrastructure they control.
Third, the coalition challenges the idea that closed AI systems are inherently safer.
While acknowledging that open models cannot be recalled once released, the letter argues that concentrating advanced AI within a handful of providers creates its own risks. Drawing on long-standing cybersecurity principles, the signatories contend that transparency enables broader testing, research and defensive security improvements rather than relying on secrecy alone.
The letter also addresses model distillation, defending it as a standard machine learning technique while distinguishing legitimate research from the unlawful extraction of proprietary model value.
A Policy Debate Before New Regulation
The letter is not tied to any specific legislation.
Instead, it serves as a public statement ahead of expected federal AI policy discussions, urging lawmakers to expand access to compute resources, support shared datasets and evaluation tools, and avoid what it describes as premature restrictions on open-weight AI.
Many of the companies involved also have clear commercial interests.
For infrastructure providers such as NVIDIA, IBM and Dell, wider adoption of open models ultimately drives demand for computing hardware regardless of which company develops the underlying AI.
What This Means for Miami
Miami's AI ecosystem is built largely around companies deploying AI rather than training frontier models, making this policy debate particularly relevant.
Startups across Wynwood, Brickell and the broader South Florida technology community increasingly rely on open-weight models such as Llama and Mistral to reduce costs while maintaining greater control over their AI infrastructure.
Healthcare providers, financial technology firms and other enterprises experimenting with self-hosted AI systems also have a direct interest in how federal policy evolves.
For researchers at Florida International University and the University of Miami, open-weight models remain one of the most affordable ways to develop advanced AI applications without the budgets of the world's largest technology companies.
Any future restrictions on open AI models could significantly increase costs for startups and research institutions, making access to open-weight AI an important factor in Miami's ability to continue growing as an emerging AI and technology hub.