8 risks that could slow the AI rally

Anthony Saglimbene, Chief Market Strategist – Ameriprise Financial

June 15, 2026

Tech stocks have been fueling markets higher, driven by solid earnings, growing demand and continued investment in artificial intelligence (AI) infrastructure. So far in 2026, this momentum has played a major role in helping major U.S. indexes reach new highs.

But when one theme is driving so much of the market’s strength, it raises an important question: what could slow it down? With AI playing such a central role in recent gains, even small shifts could have an outsized impact.

With that in mind, here are eight risks to the AI rally that investors may want to keep on their radar:

1. High expectations leave little room for error

NVIDIA delivered record results in May, and the stock still declined, marking its fourth consecutive post-earnings drop. If the best AI company in the world can't rally on a profit beat, it tells investors the bar for future stock gains might be shifting.

Valuations across the AI complex assume continued acceleration, and even a modest miss on revenue, margins or guidance could trigger a repricing.

2. Greater dispersion could act as a drag on performance

The AI trade is broadening, and that could translate into a bigger gap between the best-performing and worst-performing stocks, a phenomenon known as dispersion. Periods of high dispersion can drag down the performance of benchmarks more broadly and make it more challenging for investors to pick quality stocks.

Take semiconductors as an example: Growth is expanding beyond graphic processing units (GPUs) and into CPUs, networking, memory, storage and power delivery. On a positive note, that widens the opportunity set for investors, but it also means returns across the AI ecosystem are less likely to move in lockstep (possibly a negative for overall benchmark returns).

Picking the right exposure to AI companies could matter more than simply owning "AI.”

3. Capital expenditure is becoming a harder sell

AI spending commitments among the largest cloud providers are approaching $750 billion for 2026. But the spending is absorbing a larger share of operating cash flow, and investors are starting to ask a more pointed question: when will this investment cycle start returning capital, rather than just consuming it?

Q1 results were favorable, but companies need to demonstrate increasing return on investment (ROI) from AI to avoid another round of disruption, in our view.

4. Rising interest rates could pressure AI stocks

When interest rates are higher, investors may be less willing to pay premium prices for companies whose biggest growth may still be years away. Still, we believe AI fundamentals remain strong even if rates don't fully cooperate in the intermediate term.

5. Power and infrastructure remain a bottleneck

The AI buildout is increasingly constrained by the physical world. Grid access, transformer/switchgear shortages, utility interconnection delays, and local permitting timelines are all slowing data center expansion.

Overall, the next AI constraint may not be chips. It may be some of the other components around the buildout.

6. Software disruption isn’t going away

AI agents are challenging traditional software pricing models, seat-based licensing and growth assumptions across the broader Software-as-a-Service (SaaS) market. For example, Salesforce's latest revenue outlook came in slightly below analyst expectations, adding to concerns that AI could further harm the software market over time.

Simply put, the market is beginning to distinguish between software platforms that can profit from AI and those that could be displaced. This could create periods of broader volatility in the tech sector and markets as a whole.

7. Crowded positioning could amplify downside risks

The AI trade is very crowded among all types of investors, and when too many investors pile into the same position, the downside risk grows.

It’s important to remember that potential negative catalysts (e.g., macroeconomic or geopolitical factors) could trigger sharp unwinds in tech stocks from time to time, regardless of overall fundamentals.

8. Regulatory and confidence shocks

AI oversight, antitrust action, data-use restrictions, social mistrust or a high-profile AI failure could shift the market narrative quickly, in our view.

While these are low-probability issues in the near term, they could have an outsized impact in a market environment where AI excitement is overheated.

Bottom line

We believe the AI engine is still running strong, and we expect that fundamental strength to continue. In fact, the long-term outlook looks even stronger after recent earnings reports. But the external environment has changed: Interest rates are elevated, more investors have piled into Tech and expectations for future growth are very high. When the engine is running this hot, it’s worth thinking about what could cause it to overheat.

For investors, this may be a good time to review your investment strategy and ensure your portfolio remains well-diversified. An Ameriprise financial advisor can help you stay balanced, avoiding overexposure in any one theme while capturing opportunities across the market.