OpenAI is making a direct case for the kind of AI regulation it wants the U.S. government to pursue. In a new "economic blueprint," the company lays out policies it says could help the U.S. and its allies support AI development while protecting national security.
The document is also a political signal. OpenAI is asking for more federal action at a time when AI policy in the U.S. remains fragmented across states, and when the company is expanding its work with government agencies and its global affairs operation.
A federal plan for AI growth
The blueprint includes a forward from Chris Lehane, OpenAI's VP of global affairs. Its central argument is that the U.S. needs to move quickly to attract billions in funding for chips, data, energy and talent if it wants to "win on AI."
Lehane wrote: "To day, while some countries sideline AI and its economic potential, the U.S. government can pave the road for its AI industry to continue the country's global leadership in innovation while protecting national security."
OpenAI has repeatedly called for stronger U.S. government action on AI and the infrastructure needed to support it. The company argues that leaving AI regulation largely to states is not workable. In 2024 alone, state lawmakers introduced almost 700 AI-related bills, and some of those proposals conflict with others.
The blueprint points to Texas' Responsible AI Governance Act as one example, saying it imposes onerous liability requirements on developers of open source AI models. For OpenAI, the broader problem is not simply that states are acting. It is that companies building and deploying AI may face an uneven policy environment rather than one national approach.
Infrastructure sits at the center
OpenAI's preferred AI regulation agenda is tied closely to physical infrastructure. The company says the U.S. needs more capacity for data centers, power and transmission if it wants to lead in AI.
The blueprint recommends "dramatically" increased federal spending on power and data transmission. It also calls for meaningful buildout of "new energy sources," including solar, wind farms and nuclear.
That position fits with OpenAI's previous support for nuclear power projects, which the company and its AI rivals have argued are needed to meet the electricity demands of next-generation server farms. Tech giants Meta and AWS have run into snags with their nuclear efforts, though the source notes those issues were for reasons unrelated to nuclear power itself.
OpenAI CEO Sam Altman has also criticized existing federal laws such as the CHIPS Act, which was designed to revitalize the U.S. semiconductor industry by attracting domestic investment from top chipmakers. In a recent interview with Bloomberg, Altman said the CHIPS Act "[has not] been as effective as any of us hoped," and said there is "a real opportunity" for the Trump administration to "to do something much better as a follow-on."
Altman also argued that building major infrastructure in the United States has become too difficult, including power plants and data centers. His point aligns with the blueprint's broader message: AI policy, in OpenAI's view, is also energy policy, industrial policy and national security policy.
Safety, exports and national security
In the near term, OpenAI proposes a closer working relationship between AI companies and the federal government. The blueprint says the government should "develop best practices" for model deployment to protect against misuse and "streamline" the AI industry's engagement with national security agencies.
The company also wants export controls that allow models to be shared with allies while "limit[ing]" exports to "adversary nations." It argues that the government should share certain national security-related information with vendors, such as briefings on threats to the AI industry, and help vendors secure resources to evaluate models for risks.
The blueprint states: "The federal government's approach to frontier model safety and security should streamline requirements." It also says responsible exports to allies and partners would help them build their own AI ecosystems while relying on U.S. technology instead of technology funded by the Chinese Communist Party.
OpenAI already has U.S. government relationships. The company has deals with the Pentagon for cybersecurity work and related projects, and it has teamed up with defense startup Anduril to supply AI technology to systems the U.S. military uses to counter drone attacks.
Voluntary standards instead of mandates
One of the most important features of OpenAI's plan is what it does not endorse. The blueprint calls for standards that are "recognized and respected" by other nations and international bodies on behalf of the U.S. private sector. But it stops short of calling for mandatory rules or edicts.
Instead, OpenAI favors a "defined, voluntary pathway" for companies developing AI to work with the government on model evaluations, testing and information exchange. That approach resembles the Biden administration's AI executive order, which sought to establish high-level, voluntary AI safety and security standards.
That executive order also established the U.S. AI Safety Institute, a federal body that studies risks in AI systems and has partnered with companies including OpenAI to evaluate model safety. But Trump and his allies have pledged to repeal Biden's executive order, which could put the order's codification and the U.S. AI Safety Institute at risk of being undone.
The blueprint also addresses copyright, one of the most contested issues in AI development. OpenAI argues that AI developers should be able to use "publicly available information," including copyrighted content, to develop models.
OpenAI trains models on public data from across the web. It has licensing agreements with a number of platforms and publishers and offers limited ways for creators to "opt out" of model development. At the same time, the company has said it would be “impossible” to train AI models without copyrighted materials, and a number of creators have sued the company for allegedly training on their works without permission.
A policy push with business stakes
It is not yet clear which parts of OpenAI's blueprint, if any, will shape legislation. But the document shows that OpenAI wants to remain central to the debate over a unified U.S. AI policy.
The company has increased its political presence. In the first half of last year, OpenAI more than tripled its lobbying expenditures, spending $800,000 versus $260,000 in all of 2023.
OpenAI has also brought former government leaders into senior roles, including ex-Defense Department official Sasha Baker, NSA chief Paul Nakasone and Aaron Chatterji, formerly the chief economist at the Commerce Department under President Joe Biden.
As it expands its global affairs division, OpenAI has become more explicit about which AI laws it supports and opposes. It has backed Senate bills that would create a federal rule-making body for AI and provide federal scholarships for AI R&D. It has also opposed bills including California's SB 1047, arguing that such measures would stifle AI innovation and push out talent.
The blueprint is therefore more than a policy document. It is OpenAI's clearest attempt to define the AI regulation debate around national leadership, infrastructure buildout, voluntary safety processes, controlled exports and broader access to public information for training AI systems.