Setting the Stage: The AI Hype Cycle vs. Daly's Reality Check

The market is already betting on a transformative AI boom. The narrative is clear: artificial intelligence will supercharge productivity, fuel economic growth, and allow central banks to cut interest rates without sparking inflation. This "AI productivity boom" story is deeply priced into stock valuations and forward-looking economic models. Investors have bought the rumor, and the market is pricing in a rapid, positive shift.

Into this environment of high expectations stepped Federal Reserve Bank of San Francisco President Mary Daly. Her recent comments were a deliberate reset. While acknowledging AI's potential, she stated there is not much indication yet that artificial intelligence is fundamentally changing the US economy. Her message was a reality check against the hype cycle. She pointed to the historical precedent of electricity, noting that transformations take time and that sustained productivity gains followed a long period of adoption and reengineering. Daly's framing is classic expectation arbitrage: the market is looking ahead to a future where AI reshapes work, but the data lag is real.

The core question now is the gap between what's priced in and what's actually happening. The market consensus is betting on a near-term acceleration in productivity. Daly's caution suggests policymakers are digging for the early, disaggregated signs of change before they fully emerge. This creates a setup where any stumble in the promised productivity gains could trigger a sharp guidance reset. For now, the expectation gap is wide, and the Fed's stance-holding rates steady while leaving the door open for a few more cuts-reflects that tension between hype and the slow, data-driven reality of technological adoption.

The Evidence Gap: Productivity Data vs. Market Optimism

The market is betting on a future where AI reshapes the economy. Yet the current data tells a more measured story. On one side, we have recent hard numbers showing early gains. The U.S. economy saw productivity growth of 4.9% in the third quarter of 2025. That's a strong print, suggesting some of the promised efficiency gains are already flowing through. On the other side, the Federal Reserve is digging for the deeper, disaggregated information that would confirm these are the first signs of a lasting transformation.

President Mary Daly is waiting for that specific evidence. She acknowledges the potential, but her key point is one of timing and data quality. She stated that most macro-studies of productivity growth find limited evidence of a significant AI effect. This is the core expectation gap. The market is looking at the headline productivity number and extrapolating a broad AI boom. The Fed, however, is looking at the underlying drivers and seeing a lack of clear, economy-wide proof that AI is the cause.

Why the wait? Because the Fed needs to distinguish between temporary boosts and a fundamental shift. As Daly noted, it remains too early to declare the technology transformative. The early productivity gains could stem from other factors, or they could be the initial ripple from massive corporate AI investments that haven't yet reengineered entire industries. The central bank is waiting for the "disaggregated information" that would foreshadow a broader transformation-like clear patterns of AI adoption leading to sustained gains in specific sectors before they spread.

The bottom line is that the market has priced in a story of imminent, widespread AI impact. The current data shows some early, positive results but not the definitive, economy-wide evidence the Fed requires. This creates a setup where the next major productivity report will be scrutinized not just for its level, but for its composition. Any sign that gains are concentrated in a few tech-heavy areas, rather than broad-based, could reinforce Daly's caution and keep the door to rate cuts firmly closed.

Policy Implications: What Does 'Not Transformative' Mean for Rates?

The Fed's stance on rates is now squarely in the middle of a debate about AI's economic impact. On one side, the central bank is holding steady, waiting for clearer evidence that the technology is driving broad, sustainable productivity gains. On the other, a growing body of analysis suggests AI could be a powerful force for growth without sparking inflation. This tension shapes the near-term outlook.

The dovish view, championed by some analysts, sees AI as a tool for non-inflationary expansion. The recent productivity surge supports this. In the third quarter, productivity growth jumped to 4.9%, driven by output rising much faster than hours worked. This dynamic allows the economy to grow above its trend without pressuring wages or prices. As one analysis notes, faster productivity growth also helps offset structural headwinds from a shrinking workforce. In this scenario, AI adoption is the engine, and the Fed could eventually cut rates as disinflation continues.

Yet the Fed's caution is understandable. The central bank is wary of persistent inflation, noting that inflation based on personal consumption expenditures remains elevated at 3 percent. It sees risks that inflation will stay above target, which means it needs to see more evidence that goods price pressures are retreating before easing policy. The Fed's current plan is to hold rates steady for some time as we assess incoming data, a stance that reflects the expectation gap Daly highlighted.

A key factor in this debate is the neutral interest rate-the level where monetary policy is neither stimulative nor restrictive. AI could push this rate higher. Massive corporate investment in AI infrastructure, with capital spending on artificial intelligence expected to climb still higher, increases demand for capital and could support higher long-term rates. This is the "raise the neutral rate" argument. However, the current market is showing selective discipline. While capex estimates are rising, investors have rotated away from AI infrastructure companies where growth in operating earnings is under pressure. This suggests the market is already pricing in the high investment costs and questioning the immediate returns.

The bottom line is a wait-and-see policy. The Fed is not dismissing AI's potential, but it is not betting on it yet. With the labor market stabilizing but job creation near zero, and inflation still elevated, the central bank's prudent course is to assess the data. The next major test will be whether the recent productivity gains are sustained and broad-based. If they are, the dovish growth story could gain traction, and the door for rate cuts may open later this year. If not, the Fed's caution will likely persist, keeping rates higher for longer.

Investor Positioning: Where the Expectation Gap Creates Risk

The market's high expectations for AI are now creating a dangerous divergence in investor behavior. While the narrative of a transformative boom persists, a growing undercurrent of caution is emerging, particularly among the lenders who fund corporate investment. This tension sets the stage for a potential crash in AI-related assets if the promised productivity gains fail to materialize as quickly as priced in.

The risk is starkly highlighted by a recent survey of leveraged loan market participants. Nearly a quarter of lenders surveyed believe an AI investment crash is the most underestimated financial market risk for 2026. This is a significant warning from the institutions that provide capital for corporate spending, including massive AI infrastructure projects. Their caution stems from a mix of practical concerns: elevated loan defaults, fraud worries, and a clear sense that the economic outlook is weakening. In this environment, the high-stakes bets on AI are viewed as a vulnerability, not a surefire winner.

This lender skepticism contrasts with the performance of AI stocks, which have rallied on the promise of a productivity boom. The expectation gap here is the core risk. The market is pricing in a smooth, rapid transition where AI fuels growth and justifies higher valuations. The reality check from lenders is that the path is fraught with uncertainty and potential for a painful correction. If the promised efficiency gains are delayed or concentrated in a few sectors, the capital-intensive nature of AI adoption could strain corporate balance sheets, triggering the very default risk that lenders are already bracing for.

Yet, within this caution, there are overlooked opportunities. The focus on AI infrastructure companies is intense, but the real beneficiaries of a productivity surge may be the companies that use AI to improve their operations and margins. These are the "productivity beneficiaries"-firms in manufacturing, logistics, and services that deploy AI to cut costs and boost output. They are often less glamorous than the tech vendors but could see their earnings power rise as the technology is absorbed. The market's rotation away from AI infrastructure stocks where earnings growth is under pressure suggests investors are already starting to differentiate. The next move may be a rotation into these operational beneficiaries, where the link between AI investment and tangible financial results is clearer and less speculative.

The bottom line is that the expectation gap is creating a bifurcated market. On one side, a crash risk in the capital-intensive AI build-out is being priced by cautious lenders. On the other, the search for real productivity gains is shifting focus to the companies that will actually use the technology to grow. For investors, the risk is not in AI itself, but in betting on the wrong part of the ecosystem at the wrong time.

Conclusion: Navigating the Expectation Arbitrage

The setup is clear. The market has priced in a near-term, broad AI boom that will supercharge productivity and justify high valuations. The Federal Reserve, led by President Daly, is waiting for the disaggregated data that would confirm this transformation is underway. The gap between these two views is the core expectation arbitrage opportunity-and the source of significant risk.

The evidence points to a market in bubble territory, with the U.S. stock market having been in that state for a prolonged period since the December 2021 peak. History shows that such divergences from long-term trends eventually correct. AI, as a highly visible and self-evidently significant innovation, is following the classic pattern of technological euphoria, over-investment, and eventual market decline. The recent rally, sparked by ChatGPT, acted as a multi-stage rocket that halted a painful bear market but did not reset the underlying valuation excess.

For investors, the practical takeaway is one of humility and selective positioning. The timeline for AI's economic impact is longer and more uncertain than the market assumes. As Daly noted, transformations take time, and it remains too early to declare the technology transformative. This creates a bifurcated landscape. On one side, there is a crash risk in the capital-intensive AI build-out, a view shared by a quarter of leveraged loan market participants. On the other, there are operational beneficiaries-companies using AI to cut costs and boost margins-where the financial payoff may be clearer and less speculative.

The path forward requires navigating this gap. The safest play is to avoid betting on the wrong part of the ecosystem at the wrong time. This means questioning the immediate returns on massive AI infrastructure spending and rotating toward firms that demonstrate tangible productivity gains. The market's optimism is a powerful force, but history teaches that it is often the most central and critical of our biases. In the end, the expectation arbitrage is not about predicting the future of AI, but about recognizing the gap between what is priced in and what the data will eventually show.