OpenAI is making a bold call: The US government should take the reins on AI regulation, overriding more restrictive state laws. In a recent filing, the company argues that a unified federal approach would prevent unnecessary hurdles for AI innovation while keeping the country competitive on the global stage.
In a 15-page response to the federal Office of Science and Technology’s AI Action Plan consultation, OpenAI emphasized the need for a centralized regulatory framework. The company believes state-specific AI laws create inconsistencies, making it harder for developers to navigate compliance.
One of OpenAI’s biggest concerns? Keeping up with China. The company warned that strict state regulations could slow down American AI progress, giving China an advantage. Interestingly, OpenAI suggested a regulatory model similar to China’s—where AI companies work closely with the government—while maintaining the US’s innovation-driven culture.
To support AI startups and developers, OpenAI proposed creating a sandbox environment where companies could experiment with new AI technologies under federal oversight. This would include liability protections and exemptions from restrictive state laws, particularly those focused on frontier AI security.
Additionally, OpenAI asked the government to provide AI companies with classified threat intelligence to help mitigate national security risks. The goal? Strengthen AI’s role in cybersecurity while keeping the US ahead in the AI race.
One of the more controversial aspects of OpenAI’s filing is its stance on training AI models using copyrighted material. The company wants to preserve AI's ability to learn from publicly available data, even if that means bypassing copyright restrictions. This has been a hot-button issue, with many creators and rights holders arguing that AI should not be allowed to train on protected content without permission.
While OpenAI’s proposals are ambitious, they face a major hurdle: Only Congress has the power to override state laws. In a small footnote on page 6 of its filing, OpenAI admitted that the White House alone cannot enforce these changes—it will require legislative action. That means the company will need significant political backing to make its vision a reality.
Forrester Senior Analyst Alla Valente noted that OpenAI’s stance aligns with what the White House likely wants to hear—fewer state regulations and more room for AI development. However, she cautioned that reducing state laws could actually increase complexity for businesses operating globally.
Enterprise IT leaders already juggle AI compliance across multiple regions, including the EU, UK, Canada, and Australia, where laws differ widely. Simplifying US regulations wouldn’t necessarily make compliance easier worldwide.
While OpenAI’s proposal is ambitious, industry experts remain sceptical about whether it will gain traction. IDC Research VP Dave Schubmehl suggested that state-level AI regulation is likely inevitable, making OpenAI’s push for federal preemption a long shot.
Still, the debate over AI regulation is far from over. With increasing scrutiny on AI ethics, data privacy, and national security, policymakers will need to find a balance between innovation and protection. Whether OpenAI’s call for centralised AI regulation will shape future policies remains to be seen—but it’s clear the AI industry is pushing for major changes.
By clicking "Accept", you agree to the storing of cookies on your device to enhance site navigation, analyze site usage and assist in improving your experience.
