On February 6, 2025, the US National Science Foundation issued a Request for Information (RFI) on an Action Plan to execute the Trump Administration’s vision for AI research and development.

Among the organizations that responded to the RFI was Hugging Face, creator of the rapidly-growing platform that has become the go-to destination for repositories of open source machine learning models. In its response, Hugging Face proposed an Action Plan centered on three pillars:

1. Strengthen Open and Open-Source AI Ecosystems: technical innovation comes from diverse actors across institutions. Support for infrastructure like NAIRR and investment in open science and data allows these contributions to have an additive effect and accelerate robust innovation.
2. Prioritize Efficient and Reliable Adoption: spreading the benefits of the technology by facilitating its adoption along the value chain requires actors across sectors of activity to shape its development. More efficient, modular, and robust AI models require research and infrastructural investments to enable the broadest possible participation and innovation, enabling diffusion of technology across the U.S. economy.
3. Promote Security and Standards: Decades of practices in open-source software cybersecurity, information security, and standards can inform safer AI technology. Promoting traceability, disclosure, and interoperability standards will foster a more resilient and robust technology ecosystem.

In support of the first pillar, Hugging Face cites some eye-opening data:

Research has shown that open technical systems act as force multipliers for economic impact, with an estimated 2000x multiplier effect — meaning $4 billion invested in open systems could potentially generate $8 trillion in value for companies using them. These economic benefits extend to national economies as well. Without any open-source software contributions, the average country would lose 2.2% of its GDP.

Hugging Face further argues that commercial adoption of open models simply makes good business sense:

The commercial adoption of open models is driven by several practical factors. First, cost efficiency — developing AI models from scratch requires significant investment, so leveraging open foundations reduces R&D expenses. Second, customization — organizations can adapt and deploy models specifically tailored to their use cases rather than relying on one-size-fits-all solutions. Third, reduced vendor lock-in—open models give companies greater control over their technology stack and independence from single providers. Finally, open models have caught up to and in certain cases, surpassed the capabilities of closed, proprietary systems: Olympic-Coder, released as part of Hugging Face’s Open R1 project, surpasses Claude 3.7, the latest proprietary model from Anthropic, in terms of coding performance. All this is particularly valuable for startups and mid-sized companies, which can access cutting-edge technology without massive infrastructure investments. Banks, pharmaceutical companies and other industries have been adapting open models to specific market needs, demonstrating how open-source foundations support a vibrant commercial ecosystem across the value chain.

Hugging Face’s response articulates all of the reasons why open source is becoming the dominant development and licensing model in the AI landscape and the buzzword of choice among policymakers grappling with the question of how to regulate AI.