CFTC Chair Calls AI Compute the Digital Oil of the Modern Economy

Artificial intelligence is rapidly becoming more than a software story. It is increasingly a competition over computing infrastructure, semiconductor technology, electricity, data centers and access to advanced hardware. In the supplied material, CFTC Chair Michael Selig describes AI compute as “essentially a digital oil,” arguing that the United States must dominate the compute markets if it wants to win the global artificial-intelligence race. The comparison to oil is significant because computing capacity is becoming an increasingly important strategic resource. Modern AI systems require enormous amounts of processing power. Training and operating advanced models can involve large networks of specialized chips, high-speed connections, storage systems and data centers capable of consuming substantial amounts of electricity. For much of the technology industry's history, computing power was treated primarily as an economic input. The rapid development of generative AI has changed that perception. Companies and governments increasingly view access to advanced chips and large-scale computing infrastructure as a strategic advantage. The United States occupies an important position in this ecosystem, but the supply chain is global. Advanced semiconductor design, manufacturing, packaging, equipment and raw materials can involve companies operating across several countries. That creates vulnerabilities when geopolitical tensions affect trade, technology transfers or access to critical components. Selig's comments therefore connect AI policy with broader questions about economic security. If computing capacity becomes essential infrastructure for everything from financial services to scientific research, nations with greater access to advanced compute could gain significant advantages. The comparison with oil also highlights the difference between physical resources and digital infrastructure. Oil must be extracted, transported and refined before it becomes useful. AI compute similarly depends on a chain of infrastructure: semiconductor fabrication, chip design, networking equipment, data centers, electricity generation and software. Energy is particularly important. Advanced data centers require substantial and reliable electricity supplies. As AI adoption expands, the demand for power could become a major constraint. Countries competing for leadership in AI may therefore need to expand electricity generation, transmission networks and data-center capacity alongside semiconductor production. The semiconductor industry is another critical component. Advanced AI systems rely heavily on specialized processors designed to handle large computational workloads. Access to cutting-edge chips can influence how quickly companies can train models and deploy AI services. This is why governments have increasingly treated semiconductor manufacturing as a strategic priority. Building domestic capacity can reduce dependence on foreign suppliers, although semiconductor production is technically complex and requires enormous investment. The financial implications are equally important. Companies controlling valuable computing infrastructure could gain significant economic power as demand for AI services increases. Cloud providers, chip manufacturers, data-center operators and energy companies may all benefit from the expansion of AI-related demand. At the same time, concentration creates potential risks. If only a small number of companies control critical computing resources, smaller businesses and researchers could face higher costs or limited access. Policymakers may therefore have to balance investment incentives with competition concerns. The CFTC's involvement also demonstrates how AI is becoming relevant to financial and commodities regulators. Although AI compute is not literally a commodity like crude oil, the analogy reflects the growing recognition that computational capacity can have strategic economic value. AI could eventually influence almost every major industry. Financial institutions can use it for analysis, risk management and automation. Manufacturers can deploy it for production optimization. Healthcare researchers can use it to analyze complex datasets. Governments can apply it to public services and scientific research. That widespread usefulness makes compute capacity increasingly important. If AI becomes a foundational layer of the economy, access to computing power could become comparable to access to other essential infrastructure. The United States therefore faces a dual challenge. It must maintain technological leadership while ensuring that sufficient energy, chips, data centers and skilled workers exist to support continued expansion. Winning the AI race may require much more than developing the best algorithms. The digital-oil comparison captures that idea. Algorithms may receive the most attention, but advanced AI ultimately depends on physical infrastructure. Chips need factories. Data centers need electricity. Networks need equipment. And all of it requires capital. The emerging competition over AI compute could consequently reshape industrial policy for years to come. Countries that secure reliable access to advanced computing infrastructure may gain an important advantage in technology, finance, manufacturing and national security. The key question is no longer simply who can build the smartest AI model. It is increasingly about who can build, power and control the infrastructure required to run those models at global scale. If compute becomes the resource that powers the next generation of technology, the countries that secure that resource could hold one of the most valuable strategic advantages of the digital economy.