AI, Valuations and the U.S. Market: Are Investors Asking the Right Questions?

Why the U.S. sits at the centre of the debate

The structure of the U.S. equity market makes sense if it is uniquely tied to this discussion. Relative to most other regions, the U.S. is more "asset light", more software-intensive and more concentrated in platform-oriented business models. Profitability is higher and valuations are richer.

Those features can be tremendous strengths. When scale builds moats, dominant firms can widen their lead and compound value for years. But the same characteristics also increase exposure to faster competition. If AI tools become widely available, innovation cycles could compress. Capabilities that once differentiated leaders might spread more quickly across industries.  In short, the U.S. market has more to gain if AI reinforces scale, and more to lose if it erodes it.

Revenue pressure is real, but often overstated

There are clear channels through which AI could weigh on company revenues. Greater price transparency, lower switching costs and easier product replication can make it harder to defend premium pricing. These risks are particularly visible in certain areas of application software, digital services and consumer technology.

Yet broad indices are more diverse than the narrative sometimes implies. Energy producers respond to commodity markets. Utilities and Banks operate within regulatory frameworks. Industrial and materials companies depend on infrastructure cycles and physical capacity. Even inside technology, the largest firms often benefit from powerful ecosystems, distribution reach and embedded customer relationships that are not easily duplicated simply because a tool becomes cheaper.

For many businesses, AI may change how they operate more than it changes what they can charge.

Margin normalization does not equal earnings collapse

Because AI can lower production costs, it is reasonable to expect some benefits will flow to consumers, particularly in competitive industries. Given that U.S. corporate margins remain high relative to long-term trends, some giveback would not be surprising.

But investors should keep the scale of potential change in perspective. A gradual decline of 1% or 2% in operating margins spreads over several years would slow earnings growth, not derail it. On its own, that outcome rarely warrants a dramatic underweight shift in strategic asset allocation. History offers many examples where margins moved without permanently impairing equity markets.

Valuation duration is the more powerful force

Where AI could have a larger impact is in how markets think about the "durability" of excess returns. High-multiple companies trade at premiums because investors assume they can defend their advantages far into the future. If technological change accelerates disruption or increases uncertainty around those advantages, the price investors are willing to pay for each dollar of earnings may fall.  And those adjustments can be significant.

For example, a move from roughly 20 times earnings to 16 times earnings over a five year period could meaningfully reduce annual returns by roughly 4% to 5% even if profit growth remained intact. Compared with moderate margin pressure, changes in valuation expectations are often the heavier weight.

Looking abroad provides perspective

Other regions are not immune to these forces, but they start from different places. Many developed international markets operate with lower margins and cheaper valuations, which may naturally buffer them from multiple compression. Moreover, their businesses tend not to be asset light, with plenty of opportunity to use AI to improve business processes which all else being equal will flow to improved earnings.

Canada's equity market, with its heavier exposure to financials, materials and energy, often experiences AI more indirectly and can benefit from steadier income characteristics.  Emerging markets may offer lower entry prices still, though typically with greater volatility in outcomes. Emerging markets are particularly interesting because they are suppliers of much of the physical goods required for the ongoing AI investments being made around the world.

Relative positioning, therefore, depends as much on starting valuation as on technological promise.

What would change the allocation case

To argue for a decisive structural reduction in U.S exposure, we would need evidence that goes beyond theoretical risk. Sustained commoditization of software pricing, regulatory constraints that materially limit monetization, or a clear and lasting repricing relative to global peers would be more convincing signals. Without such developments, the growth capacity and innovative strength of U.S. companies can counterbalance moderate competitive pressure.

If AI ends up reinforcing the dominance of the largest ecosystems, the U.S.  could continue to lead global returns. If value creation becomes more evenly distributed, relative performance may soften but not collapse. Only in a more severe reassessment, where markets broadly conclude that competitive advantages fade much faster than previously believed, would we expect widespread valuation drag. In that environment, higher-multiple markets would simply feel it more.

For investors, the implication is not retreat but selectivity. Exposure to companies with durable platforms and proven adaptability remains important. At the same time, diversification across regions can provide ballast if valuation assumptions evolve. The goal is balance: participating in innovation while recognizing the price paid for it.

The bottom line

AI is transformative. It will reshape industries, alter costs curves and influence how firms compete. But transformation does not automatically mean deterioration. The more immediate risk for U.S. equities is not widespread earnings destruction; it is the possibility that investors revisit how long exceptional profitability can endure.

Maintaining valuation discipline and managing concentration may therefore matter more than reacting to the latest AI headline.