US stocks are overvalued because of unrealistic expectations for AI-powered economic growth, Vanguard says

• Investors are too optimistic about the near-term prospects of AI, Vanguard said.
• Firms would need to growth profit by 40% annually for the next three years to match valuations, the firm said.
• "This is double the annualized rate of the 1920s, when electricity lit up the nation," Vanguard wrote.

With tech companies still pushing the boundaries of artificial intelligence, market excitement for it seems endless.

But this enthusiasm expects too much from the technology in too little time, Vanguard wrote on Thursday.

Wall Street is rife with upbeat forecasts about what AI could do to the economy and corporate profits. Most of them are pinned to a US workplace revolution and a productivity boom.

That optimism has helped fuel strong stock gains, with the benchmark S&P 500 up 18% year-to-date through Thursday.

But Vanguard global chief economist Joe Davis thinks expectations are too high, and says that stocks are overvalued even if the AI boom plays out as anticipated.

He estimates that US corporate profits would have to growth by 40% annually over the next three years to justify where stocks are trading now. For context, the S&P 500's trailing one-year earnings growth rate through the second quarter of 2024 was 10.9%, according to FactSet data .

"I'm optimistic about the long-term potential of artificial intelligence to power big increases in worker productivity and economic growth," global chief economist Joe Davis wrote. "But I'm pessimistic that AI can justify lofty equity valuations or save us from an economic soft patch this year or next."

He continued: "This is double the annualized rate of the 1920s, when electricity lit up the nation — not to mention economic output and corporate income statements."

Such a historic surge in corporate performance looks even less probable if the economy cools down next year. Vanguard expects GDP to expand by just 1% to 1.5% in 2025.

It's not that the investment firm has no faith in AI's potential — its research suggests 45% to 55% odds that AI will trigger a boom in labor productivity. Between 2028 and 2040, that could spur a 3.1% annualized rate of US growth in real terms.

But investors need to let go of any notions that this will happen immediately, Davis said. While firms have poured billions to advance their position in the sector, some market players are incorrect in thinking that AI investing will reach $1 trillion in the near term:

"$1 trillion in AI investment by 2025 would require 286% growth. That's probably not going to happen, which means we're unlikely to experience an AI-driven economic boom in 2025," he said.

Some on Wall Street are much more pessimistic. BlackRock has said there's a strong chance that heavy AI investing will trigger higher inflation before any production boom can come. That could erode corporate profit growth.