Mozilla : Open Policy & Advocacy  -  Mozilla Respond to the White House’s RFI on AI

 -  24 mars - 

The Future of AI Must Be Open, Competitive, and Accountable

The internet has always thrived on openness, access, and broad participation. But as we enter the AI era, these core principles are at risk. A handful of dominant tech companies are positioned to control major AI systems, threatening both competition and innovation. At Mozilla, we believe AI should serve the public interest—not just corporate bottom lines.

Earlier this month, we responded to the White House Office of Science and Technology Policy’s request for input on AI policy, where we offered a roadmap for a more open and trustworthy AI future (view Mozilla’s full submission here). Here’s what we think should happen.

1. AI Policy Must Prioritize Openness, Competition, and Accountability

Right now, too much AI development stays behind closed doors. Proprietary models dominate, creating a landscape where users and developers have little insight—or control—over the AI systems shaping our digital lives. If we want AI that benefits everyone, we need strong policies that promote:

• Openness: Encouraging open-source AI development ensures transparency, security, and accessibility.
• Competition: Preventing monopolistic control keeps AI innovation dynamic and diverse.
• Accountability: Effective governance can mitigate AI’s risks while fostering responsible development.

By advancing these principles, we can build an AI ecosystem that empowers users rather than locking them into closed, corporate-controlled systems.

2. The Government Should Support Public AI Infrastructure

AI’s future shouldn’t be dictated solely by private companies. Public investment is key to ensuring broad access to AI tools and research. We support initiatives like the National AI Research Resource (NAIRR), which would provide universities, researchers, and small businesses with AI computing power and resources. We hope to see federal, state, and local governments increasingly adopt open source AI models in their workflows, which can help save taxpayers money, increase efficiency, and prevent vendor lock-in. Public AI infrastructure levels the playing field, allowing more voices to shape AI’s future and facilitating innovation across America, not just in a few tech hubs.

3. Open-Source AI Should Be Encouraged, Not Restricted

Discussions about restricting open-source AI through export controls often miss the point about how to ensure national leadership in AI. Open-source AI fosters innovation, improves security, and lowers costs—critical benefits for businesses, researchers, and everyday users around the world. For the United States, promoting open source AI means more global products would be built on top of American AI innovation.

A 2025 McKinsey report, “Open source in the age of AI,” created in collaboration with Mozilla, found that 60% of decision-makers reported lower implementation costs with open-source AI compared to proprietary tools. Restricting open models would stifle progress and put the U.S. at a competitive disadvantage. Instead, we urge policymakers to support the open-source AI ecosystem and resist governance approaches that restrict AI models overall rather than making more targeted and precise interventions for AI harms.

4. AI Energy Consumption Needs Transparency

AI systems consume enormous amounts of energy, and this demand is only growing. To prevent AI from straining our power grids and driving up costs, we need better transparency into AI’s resource consumption so that we can plan infrastructure development more effectively. The federal government should work with the industry to collect and share data on AI energy use. By understanding AI’s impact on infrastructure, we can promote sustainable innovation.

5. The U.S. Must Invest in AI Talent Development

AI leadership isn’t just about technology—it’s about people. To remain competitive, the U.S. needs a strong, diverse workforce of AI researchers and practitioners. That means investing in:

• Community colleges and public universities to train the next generation of AI professionals.
• Apprenticeship and retraining programs to help workers adapt to AI-driven industries and adopt AI in every type of business across the economy from manufacturing to retail.
• Public-private partnerships that create novel education pathways for students, like Dakota State University’s collaboration with ArmyCyber.

By growing the AI talent ecosystem, we ensure that AI works for people—not the other way around.

The Path Forward

AI is one of the most transformative technologies of our time. But without strong policies, it risks becoming yet another tool for big tech consolidation and unchecked corporate power.

At Mozilla, we believe in an AI future that is open, competitive, and accountable. We call on policymakers to take bold steps—supporting open-source AI, investing in public infrastructure, and fostering fair competition—to ensure AI works for everyone.

The post Mozilla Respond to the White House’s RFI on AI appeared first on Open Policy & Advocacy.

par Joel Burke and Jenn Taylor Hodges