Yardeni Research is cautioning that the relentless rise in profit projections tied to artificial intelligence has created an unforgiving backdrop for investors, with AI earnings hype now so elevated that even solid results may not satisfy the market.
In a research note, the firm argued that consensus estimates for corporate earnings growth have climbed to levels that strain credibility, leaving little cushion for companies that fall short. The warning arrives as enthusiasm for AI continues to underpin long-term profit expectations across the S&P 500.
Forecasts Stretch Beyond Believable
Analysts now expect S&P 500 operating earnings per share of $366 in 2026, $425 in 2027 and $494 in 2028. Yardeni’s own projections are more conservative at the back end: $375, $425 and $450, respectively, signaling that the firm sees greater risk in later years.
The gap between consensus and Yardeni’s view is modest for 2026 and 2027 but widens sharply for 2028, when analysts foresee nearly $44 more in per-share profit than the firm does. That divergence underscores how much of the optimism is back-loaded into a period when current visibility is limited.
Longer-term growth expectations have also reached a record. The average five-year earnings growth estimate now stands at 25.6%, with technology and semiconductor companies carrying far heavier burdens. Information technology long-term growth expectations sit at 41.9%, while semiconductor firms face projections of 66%.
Valuations Cool as Expectations Heat Up
Despite the rising profit forecasts, the S&P 500’s forward price-to-earnings ratio has fallen to 18.5 from above 22 earlier in the year. That combination—higher expected earnings but lower willingness to pay for them—suggests investors are becoming more discriminating about which growth stories deserve premium valuations.
Yardeni described the situation as an extraordinarily high bar that makes the market quick to punish disappointment. Companies that report good numbers may still face selling pressure if those results do not match the lofty expectations baked into analyst models.
The firm’s caution echoes a broader tension in the market: AI-related spending and revenue are growing, but the pace required to justify current projections is steep. When forecasts outrun what investors can credibly underwrite, the cost of missing them rises.
Guardrails May Prevent an AI Bubble
Yet Yardeni does not expect the AI boom to spiral into an unchecked bubble. The firm pointed to financial discipline, political restrictions, social concerns and physical infrastructure constraints as forces that could temper excessive speculation.
Those guardrails, in Yardeni’s view, are numerous enough to prevent the AI investment cycle from running completely wild. Instead, they may push investors to focus more intently on profitability and the actual financial returns generated by large-scale AI spending.
That shift would be healthy for the market over time, even if it produces more volatility in the near term. Greater scrutiny can separate companies with durable earnings power from those riding a wave of enthusiasm without the fundamentals to support it.
Economic Strength Complicates the Fed’s Path
The earnings debate is unfolding against a resilient US economy. Real gross domestic product expanded at a 3.4% annualized rate in the third quarter, while the core personal consumption expenditures price index rose 3.1% year over year, indicating persistent inflation.
Yardeni maintains that the Federal Reserve should return monetary policy to a neutral setting, which it estimates at a range of 4.25% to 4.50%. Strong growth and sticky inflation together make the central bank’s task more delicate, and they also complicate the outlook for equity valuations.
For investors, the message is not that AI is a passing fad but that the margin for error has narrowed. Companies will need to show that their earnings can grow fast enough to justify the projections now embedded in the market—and the market, increasingly, is less willing to wait.
