At their December meeting, Federal Reserve officials estimated that the U.S. economy grew 1.7% in 2025, a year plagued by uncertainty around costly new tariffs. The forecast for 2026 growth was raised to 2.3% — a significant jump from the 1.8% rate previously forecasted in September. In a post-meeting press conference, Chairman Powell acknowledged that the increase was partly due to surging private investment in data centers and other business spending related to artificial intelligence (AI).¹

The official U.S. gross domestic product (GDP) report covering all of 2025 may not be available until late February, as the Q4 advance estimate was postponed due to the shutdown. Inflation-adjusted (real) GDP grew at an annual rate of 4.3% in Q3, the fastest pace in two years.²

Nevertheless, if you are an investor, a business owner, or perhaps an employee wondering about your career prospects, you might be more concerned about what lies ahead for the economy in 2026.

Economic projections, whether they are published by the Fed or any other organization, are essentially educated guesses. Economists in the public and private sectors are tasked with trying to predict the future based on a wide range of indicators, potential risks, and their overall impressions of market conditions. So far, forecasts for 2026 suggest the new year is being met with a mix of optimism and wariness about how investments in AI could reshape the U.S. and global economies.

Fed policies and projections

The Federal Open Market Committee (FOMC), which is the arm of the Federal Reserve responsible for setting U.S. monetary policy, meets eight times per year to review economic data pertaining to inflation, employment, and credit conditions before voting on whether to adjust the benchmark federal funds rate. The FOMC typically raises rates to combat inflation and lowers rates to stimulate employment. But how should the Fed act when both measures are moving in the wrong direction? This was the situation in the latter months of 2025, when job growth was slowing and inflation was rising, partly due to tariffs on imported goods.

The federal funds rate was reduced by a quarter point after meetings in September, October, and December of 2025, ending the year in the range of 3.50% to 3.75%. These actions suggest that labor market risks appeared to outweigh the threat of inflation.³

The Personal Consumption Expenditures (PCE) price index for September 2025, released in December after a delay, indicated that inflation ticked up to 2.8%.⁴ Based on the median projection, the FOMC expects PCE inflation to slow over the coming months and end 2026 at 2.4%. This would still be higher than the Fed’s established 2.0% long-run target.⁵

The U.S. unemployment rate increased from 4.0% in January 2025 to 4.6% in November 2025.⁶ The Fed’s median projection for the unemployment rate at the end of 2026 is 4.4%, indicating that Fed members do not expect large-scale layoffs.⁷

Even so, many job seekers are being shut out of a “low-hire, low-fire” labor market, which the Fed has factored into growth forecasts. In a press conference following the December meeting, Chairman Powell said, “It’s a complicated, unusual, and difficult situation, where the labor market is also under pressure, where job creation may actually be negative.”⁸

AI investment

It’s been reported that just four tech giants — Alphabet (Google), Amazon, Meta, and Microsoft — had capital expenditures of $337.8 billion in 2025, primarily to build out AI infrastructure.⁹ By one estimate, business investment in data centers and related AI infrastructure could contribute as much as 0.5% to U.S. economic growth in 2025 and 2026.¹⁰

Data centers are home to millions of servers running 24/7 to process AI applications. In a race for competitive advantage, the “hyperscalers” that provide cloud services, along with other companies aiming to profit from the AI boom, are investing in AI infrastructure at a furious pace. As of now, there isn’t enough computing power in the world — the hardware, processors, memory, storage, and energy needed to operate data centers — to fulfill AI demand.

If current demand trends continue, $5.2 trillion in global AI-driven capital expenditures would be required by 2030, according to calculations by McKinsey & Company. But future demand is highly uncertain, as is the prospective return on investment, or ROI, for big spenders.¹¹

Global implications

According to the International Monetary Fund’s World Economic Outlook, global economic growth is projected to slow from 3.3% in 2024 and 3.2% in 2025 to 3.1% in 2026. The expected slowdown reflects headwinds from uncertainty and trade protectionism, but is an improvement from the WEO’s previous forecasts, as the tariff shock so far has been smaller than originally anticipated. U.S. growth of 2.0% is expected in 2026.¹²

Recent forecasts from S&P Global are similar, predicting global economic growth of 3.2% and U.S. growth of 2.0%. However, economists surmise that an AI-related investment boom is boosting U.S. GDP growth and “masking general economic softness.”¹³

Both forecasts highlighted AI investments as a growth driver and noted the potential for economy-wide gains if AI speeds up productivity growth as much as expected. On the other hand, both also considered the possibility of an abrupt repricing of AI stocks, which could cause a pullback in AI investments, to be a downside risk to their growth forecasts. Other downside risks are related to trade policy unpredictability, labor demand, and higher bond yields.^14–15

At a December press conference, Fed Chair Powell was asked whether AI is a factor in recent job market weakness. His response speaks to the uncertainty that many Americans feel as they enter this pivotal new year: “It’s probably part of the story. It’s not a big part of the story yet. We don’t know whether it will be.”¹⁶

Forecasts are based on current conditions, are subject to change, and may not come to pass. All investments are subject to market fluctuation, risk, and loss of principal. When sold, investments may be worth more or less than their original cost.