Memo Published June 17, 2026 · 4 minute read
How Will AI Impact the Federal Budget?
When The Terminator premiered in 1984, the national debt was $1.6 trillion, roughly 40% of GDP. Today, debt exceeds the size of the US economy, while artificial intelligence has moved from science fiction into the real economy. That has led some to wonder: can AI improve America’s fiscal outlook?
In our view, the answer is “yes…but.” AI could increase productivity and federal revenues while helping the federal government operate more efficiently. But those gains will take time, come with upfront costs, and won’t come close to eradicating our budget deficit. In this memo, we explore how AI could affect the federal budget, what experts estimate, and why AI is likely to be a helpful tailwind rather than a budgetary rescue plan.
How Can AI Impact the Federal Budget?
AI could affect the federal budget in two main ways: by changing the broader economy and by changing how the federal government administers programs. Those effects could increase or decrease revenues and spending. The Congressional Budget Office (CBO) emphasizes that the timing, size, and direction of AI’s impact on the federal budget is uncertain.1
Changing the Broader Economy
If AI helps businesses become more productive, the economy could grow faster, incomes could rise, and the federal tax base could expand. But the fiscal effects are complicated. In the short term, businesses may reduce taxable income by deducting the cost of AI investments. Over time, AI could impact who earns income and how that income is taxed. For example, if AI increases returns to business owners and investors while reducing workers’ wages or employment, some income could shift away from labor compensation and toward capital income, which is taxed differently. On the other hand, if AI makes workers more productive, creates new tasks, or supports new industries, it could raise wages and increase tax receipts.2 In addition, stronger economic growth and productivity generally leads to higher interest rates.3
Changing Government Operations
AI could also identify fraud, waste, and improper payments in government programs before money goes out the door rather than after the fact. It could also strengthen tax administration by helping the IRS target audits, improve compliance assistance, and reduce the tax gap. These efforts could produce meaningful savings or additional revenue if they are implemented well.4
But AI has costs too. The federal government would need to make upfront investments in technology, data quality, staff training, privacy protections, and human oversight before those gains materialize. Poor data, weak safeguards, or over-reliance on automated systems could lead to mistakes, bias, delayed benefits, or missed fraud.5
How Big Will the Impact Be?
Key forecasts suggest AI will provide a measurable, but modest, fiscal benefit.
CBO expects AI to modestly increase productivity and economic growth over the next decade, helping raise GDP and federal revenues.6 Estimates from the Penn Wharton Budget Model similarly suggest AI could provide a small but persistent boost to productivity growth over time.7
Using those estimates, AI could reduce projected federal debt by roughly $435 billion by 2035 while increasing real GDP by about 1.9%.8 But the total budget effect remains limited: annual deficits would remain large, shrinking very modestly (down 0.3 percentage points of GDP by 2035) compared to a scenario without AI.9
Some estimates are more optimistic. The Budget Lab at Yale, using surveys of AI and economics experts rather than its own assessment, finds AI could reduce deficits more substantially (down 1.6 percentage points of GDP) under favorable productivity scenarios. But those fiscal improvements shrink considerably (to 0.8 percentage points of GDP) once the model accounts for potential labor market disruptions and spending tied to displaced workers.10
Overall the impact on the fiscal outlook is likely to be limited. The biggest drivers of long-term debt—an aging population, rising health care costs, higher interest payments, and slower labor force growth—are much larger and more predictable than the budgetary gains most experts expect from AI.11
AI adoption will also take time. Many businesses and agencies will need years to test tools, update systems, train workers, and figure out where the technology actually delivers value. That means any boost to productivity, revenues, or program integrity is likely to build gradually rather than appear all at once.12 AI could outperform current forecasts. But if major productivity gains also trigger widespread job destruction rather than labor market disruption, the net fiscal benefits may not materialize as all.
Conclusion
Amid all the AI uncertainty, experts agree on one thing: AI adoption will likely improve productivity and modestly strengthen economic growth. Those gains could marginally improve the fiscal outlook through higher revenues and more efficient government operations. But AI is unlikely to meaningfully alter the trajectory of the national debt, which is driven primarily by structural demographic and spending pressures. Tough policy decisions, not technological advancements, are the only path to a sustainable fiscal future.
