The AI Rally Is Missing One Key Warning Sign

One of the most persistent questions facing investors today is whether the excitement surrounding artificial intelligence resembles the technology bubble of the late 1990s.

The comparison is understandable. A transformative technology is reshaping business models, capital has concentrated in a relatively small group of companies, and market leadership has narrowed in ways that feel familiar to anyone who lived through that earlier period.

But markets are rarely defined by headlines or analogies alone. The more important differences tend to appear beneath the surface, particularly in how risk develops and how volatility behaves when prices are tested.

In the late 1990s, many of the companies driving the technology boom were unprofitable, highly leveraged, and dependent on aggressive growth assumptions. Capital spending surged far ahead of demand, leaving little room for error. When expectations shifted, volatility did not simply rise—it remained elevated as selling pressure fed on structural weakness and balance sheet stress.

Today’s market looks different in several important respects.

Many of the companies most closely associated with artificial intelligence generate substantial cash flow, maintain stronger balance sheets, and fund investment internally. That does not make markets immune to corrections, but it does influence how risk tends to surface.

In environments supported by profitability rather than leverage, declines are more often driven by repositioning than distress, and volatility is more likely to normalize rather than persist.

This distinction matters because volatility often provides information before price does. Periods of unusually high volatility tend to coincide with fear, uncertainty, or emotional decision making. Just as important: the resolution of that volatility frequently precedes meaningful changes in market behavior. Whether markets stabilize, continue higher, or transition into something more challenging, those shifts are typically reflected in volatility first.

That insight led me years ago to focus my research on volatility itself rather than on prediction or narrative. Instead of trying to forecast outcomes, I developed a framework designed to measure how markets are behaving in real time by analyzing volatility at the individual stock and ETF level. By doing so, it becomes possible to identify when fear is elevated, when it begins to subside, and when risk is changing in ways that matter.

Because volatility adapts as market conditions evolve, this approach does not depend on a specific environment to work. Bull markets, bear markets, and periods of uncertainty all express risk differently, but volatility reflects those changes without requiring a shift in process. The framework remains consistent even as the market does not.

The technology bubble of the late 1990s produced extraordinary gains, but it was also defined by accelerating instability. Volatility did not merely spike. It stayed elevated as the market moved into a prolonged period of stress.

So far, the current AI driven advance has behaved differently. While pullbacks have occurred and uncertainty has surfaced at times, volatility has generally remained contained relative to past bubble periods. That suggests a market still digesting change rather than one unraveling under its own weight.

Markets will continue to evolve, and no environment lasts forever. But meaningful regime shifts rarely arrive without warning. They tend to leave traces in volatility behavior well before they become obvious in price action or headlines. That is why a disciplined approach does not rely on guessing which story will dominate next. It relies on observing how markets respond as those stories unfold.

Narratives will change. Technologies will advance. Comparisons will come and go. A process grounded in volatility allows investors to navigate those shifts without relying on emotion or prediction. The goal is not to forecast the future. It is to manage risk, recognize opportunity, and respond to what the market is actually doing.

The current enthusiasm around artificial intelligence may echo aspects of past technology cycles, but it does not yet resemble the late 1990s in the way risk is expressing itself. Strong profitability, lower leverage, and clearer paths to monetization have helped keep volatility contained. When that changes, volatility will reflect it, as it always does.

Until then, the most reliable guide is not the loudest headline or the most familiar analogy, but the behavior of the market itself.