OpenAI is making a more detailed case for federal support around the infrastructure it says is needed to build and run large-scale AI systems in the US.
In a recent letter from chief global affairs officer Chris Lehane to Michael Kratsios, the White House’s director of science and technology policy, the company urged the government to consider widening an existing Chips Act tax credit so it reaches more parts of the AI infrastructure stack.
What OpenAI Asked For
The central request concerns the Advanced Manufacturing Investment Credit, known as AMIC. The source describes AMIC as a 35% tax credit included in the Biden administration’s Chips Act.
OpenAI argued that the credit should not be limited to semiconductor fabrication. Its letter said the government should consider expanding the credit to include electrical grid components, AI servers, and AI data centers.
That framing matters because AI data centers are not just buildings with computers inside. Based on the company’s request, OpenAI is treating data center construction, power infrastructure, servers, and semiconductor capacity as connected pieces of the same buildout.
Lehane wrote that expanding AMIC coverage would “lower the effective cost of capital, de-risk early investment, and unlock private capital to help alleviate bottlenecks and accelerate the AI build in the US.”
Permits, Reviews, and Raw Materials
The tax credit was not the only item in OpenAI’s letter. The company also called for the government to accelerate permitting and environmental review for these projects.
That request points to a practical bottleneck: even when private capital is available, major infrastructure projects can be slowed by approval processes. OpenAI’s position, as described in the source, is that faster review would help speed the construction of AI infrastructure.
The company also asked for a strategic reserve of raw materials needed to build AI infrastructure. The materials named in the source include copper, aluminum, and processed rare earth minerals.
Taken together, the requests show a broad view of what AI expansion requires. OpenAI is not only discussing models or chips. It is also asking policymakers to think about power equipment, servers, data centers, materials, and review processes as part of the same national infrastructure question.
Why the Letter Drew Fresh Attention
OpenAI first published the letter on October 27. According to the source, it did not receive much press attention until later, when comments from company executives triggered a wider discussion about what OpenAI wants from the Trump administration.
At a Wall Street Journal event on Wednesday, CFO Sarah Friar said the government should “backstop” OpenAI’s infrastructure loans. She later posted on LinkedIn that she misspoke.
“OpenAI is not seeking a government backstop for our infrastructure commitments. I used the word ‘backstop’ and it muddied the point.”
CEO Sam Altman also addressed the issue. He wrote that OpenAI does not “have or want government guarantees for OpenAI datacenters.”
Altman’s post drew a line between government support for broader infrastructure priorities and direct guarantees for OpenAI’s own data centers. He wrote that governments should not “pick winners or losers,” and that taxpayers should not bail out companies that make poor business choices or lose in the market.
At the same time, Altman said OpenAI had discussed loan guarantees in a different context: supporting the buildout of semiconductor fabs in the US.
The Scale Behind the Policy Push
The size of OpenAI’s infrastructure ambitions helps explain why policy tools such as AMIC, permitting, environmental review, and raw material reserves are part of the discussion.
In the same post cited by the source, Altman wrote that the company expects to end 2025 “above $20 billion in annualized revenue run rate and grow to hundreds of billion by 2030.” He also said OpenAI has made $1.4 trillion in capital commitments for the next eight years.
Those figures place the company’s policy requests in a larger financial context. If OpenAI is planning around that level of capital commitment, then the cost of financing, the availability of components, and the pace of construction become central issues rather than secondary details.
The company’s argument is that expanding AMIC could make early AI infrastructure investment less risky and help draw in private capital. The source does not say the government has accepted that argument. It shows that OpenAI is actively asking federal officials to consider a broader definition of what qualifies for support under a major manufacturing tax credit.
What Is at Stake
The debate is likely to turn on where policymakers draw the line between enabling infrastructure and subsidizing specific corporate expansion.
OpenAI’s letter frames AI data centers, AI servers, grid components, and raw materials as infrastructure needed for the US AI buildout. Its executives, meanwhile, have tried to clarify that the company is not asking taxpayers to guarantee OpenAI data center commitments.
That distinction is important. A broader tax credit for categories of infrastructure is different from a direct guarantee for one company’s loans. The source shows OpenAI trying to make that distinction while still pressing for federal action that could lower costs and reduce friction across the AI buildout.
For now, the key fact is simple: OpenAI wants the Trump administration to consider expanding a Chips Act tax credit beyond semiconductor fabrication. If adopted, that approach would place AI data centers and related infrastructure closer to the center of US industrial policy for AI.