OpenAI CEO Sam Altman is calling on the U.S. government to expand the CHIPS and Science Act’s Advanced Manufacturing Investment Credit (AMIC) to include AI data centers, server manufacturing, and grid infrastructure.

Under current law, AMIC targets semiconductor production. Altman argues that the next wave of economic growth will hinge on compute manufacturing and energy capacity, not just chips. OpenAI’s forecasted $1.4 trillion, eight-year infrastructure build-out would require parallel expansion in tax-credit eligibility.

If Washington acts, it could lower capital costs for hyperscale AI facilities and accelerate the U.S. race with China in compute power.
If it doesn’t, the risk is fragmented investment, where global players seek friendlier fiscal climates.

Altman was careful to note that OpenAI is not seeking government loan guarantees — signaling that this is not a bailout, but a strategic tax modernization plea.

Why it matters:
Tax policy is now an AI-era industrial lever. How lawmakers respond will determine whether U.S. AI capacity scales sustainably or bottlenecks under capital constraints.

How AI Uses Our Water: When Machines Get Thirst: Cooling Systems, Data Centres, and the Infrastructure Behind Artificial Intelligence

How AI Uses Our Water: When Machines Get Thirst: Cooling Systems, Data Centres, and the Infrastructure Behind Artificial Intelligence

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Silverstone Technology RM4A 4U rackmount Server Chassis with Enhanced 360mm radiators Compatibility, SST-RM4A

Silverstone Technology RM4A 4U rackmount Server Chassis with Enhanced 360mm radiators Compatibility, SST-RM4A

Supports up to SSI-EEB motherboards

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Advanced Concepts for Renewable Energy Supply of Data Centres

Advanced Concepts for Renewable Energy Supply of Data Centres

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

The Sales Superlift: How to Win More Equipment Sales with AI as Your Side-Kick

The Sales Superlift: How to Win More Equipment Sales with AI as Your Side-Kick

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

You May Also Like

Architecting an Efficient Inference Stack: From Models to Serving

Discover how to design a streamlined inference stack that maximizes performance and reliability—continue reading to unlock the secrets of seamless deployment.

Limits, Levers, and a Roadmap: What It Will Take for Video Models to Become Vision Foundation Models

Structured abstract Background. The paper makes a bold claim—that video models are…

KV Cache Offloading: Techniques, Trade‑offs, and Hardware Support

Learn how offloading KV cache tasks with specialized hardware can enhance performance but involves critical trade-offs worth exploring.

QAT Vs Post‑Training Quantization: When to Use Which

Keen to optimize model deployment? Discover when to choose QAT versus post-training quantization for best results.