U.S. stocks are rebounding as investors buy the dip after a tech-led selloff, but artificial intelligence could be reshaping the foundations of market leadership.
BNN Bloomberg spoke with Tim Murray, capital markets strategist in the multi-asset division at T. Rowe Price, who said mega-cap tech is moving from a low-competition, asset-light model to one defined by heavy capital spending and intensifying rivalry.
Key Takeaways
- Markets are broadly rational but more volatile, with momentum-driven capital amplifying price swings beyond what fundamentals alone might justify.
- Mega-cap tech’s 15-year run was built on low capital intensity and limited competition, conditions that are now shifting as AI reshapes the landscape.
- The five largest hyperscalers are projected to spend $1.4 trillion in capital expenditures between 2026 and 2027, marking a sharp rise in capital intensity.
- Intensifying AI competition now includes both established tech giants and well-capitalized private entrants expected to tap public markets.
- Industrials, materials and energy may benefit from both AI infrastructure demand and a broader rebound in U.S. economic activity.
Read the full transcript below:
ANDREW: Well, let’s get more on the U.S. equity markets, in particular what our guest calls the possibility of a regime change in artificial intelligence and technology. He argues that AI fundamentally alters the landscape for tech companies. We’re joined by Tim Murray, capital markets strategist in the multi-asset division at T. Rowe Price. Tim, thank you very much for joining us.
TIM: Thank you for having me.
ANDREW: What do you mean by regime change in the tech market?
TIM: So I think it’s important to understand we had 15 years where mega-cap tech companies were fantastic investments, but there were two really important reasons for that: low capital intensity and low competition. Mega-cap tech companies basically dominated three industries. We had disruption in three different industries. Those industries were media — the switch from broadcast media to digital, interactive, on-demand media. We also had the switch from enterprise computing being on premises to now being in the cloud. And we also had the big switch in retail, moving from brick and mortar to online.
The mega-cap tech companies dominated those three industries, and they did so without having to spend a ton of capex. They also did so without competing a lot with each other — they kind of stayed in their own lanes. So first of all, the capex was asset-light, soft capex, and it was mostly code, intellectual property and research and development. And then on competition, they were all able to find their own lane. You had Facebook dominating the digital media shift. You had Google dominating search. You had Amazon and Microsoft dominating enterprise computing and the switch to the cloud, and Amazon dominating retail.
As a result, they had this great 10- to 15-year period where they were growing revenues dramatically and expanding margins constantly. To see companies of their size do that for such a long period of time is almost unheard of. But now we’re getting to the point where, because of AI, those two dynamics are changing dramatically.
First of all, the capex. We all know about how large the capex has been among the five big hyperscalers. They’re no longer just doing code. They’re building massive data centres and creating the power infrastructure to run them. The five largest hyperscalers are projected to spend $1.4 trillion on capex between 2026 and 2027, and those numbers keep getting revised higher. So it’s a tremendous amount of capex.
Meanwhile, they’re all saying they want to dominate AI and are not willing to cede the competition. They all want to have the best LLM. On top of that, you have new entrants. We likely have three very big IPOs coming in the U.S. market: OpenAI, Anthropic and SpaceX, which recently combined with xAI. So in addition to competing more with each other than they ever have, you now have new entrants with huge amounts of capital behind them, about to get a further infusion from the public markets. That means big changes in the dynamics on both the capital intensity and competition side.
ANDREW: What would your take be, though? I know your overall feeling is that it’s going to be much tougher for these tech giants to produce the same kind of gains over the next 15 years. Broadly, how would you pick your tech stocks now?
TIM: I think it means you need to recognize it’s going to be a much tougher environment for these companies. We know these are great companies. If you were to pick the most likely winners, these are still probably the companies that will come out ahead in AI and AI distribution five years from now. The question is how much pain they will have to endure in the next two to three years.
From a tech standpoint, I’d be concerned, especially if I’m not doing it with active management and am just doing it passively — having a big slug of my assets in tech — given that the biggest companies are facing the toughest environment they’ve ever faced.
ANDREW: Right. They’re off their highs, but those big tech stocks have generally been great investments.
TIM: Yes, and when we’ve seen pullbacks in these stocks, “buy the dip” has been the way to go for the last 10 to 15 years. What I’m saying is this time might be different. It might not be as simple as buying the dip. You could still see great returns over a five-year period, maybe even over three years, but in the near term, you want to be more selective about tech investing. You probably don’t want to blindly have a massive allocation to tech versus the rest of the market.
ANDREW: And of course, if you’re buying an index fund, by default you end up with a lot of tech. We’re tight for time, Tim, but what areas of the market are you more interested in or overweighting right now?
TIM: The big three for me are industrials, materials and energy. Because of the AI buildout, there’s a lot of demand for energy, materials and select industrial areas. We’re also seeing a broadening of the U.S. economy. In late 2024 and into 2025, the economy was almost entirely driven by AI buildout. Now we’re seeing broader economic growth.
We just got a PMI reading above 50 — only the third time in 39 months — and it jumped to 52. New orders rose from 47 to 57. There are reasons to believe we’re finally seeing a real uptick in the broader U.S. economy. That benefits industrials, materials and energy, on top of the AI tailwinds already in place.
ANDREW: It’s interesting. U.S. housing starts just hit a five-month high, and that bodes well for materials demand. Tim, thank you very much indeed.
TIM: My pleasure. Thank you.
ANDREW: Tim Murray, capital markets strategist in the multi-asset division at T. Rowe Price.
---
This BNN Bloomberg summary and transcript of the Feb. 18, 2026 interview with Tim Murray are published with the assistance of AI. Original research, interview questions and added context was created by BNN Bloomberg journalists. An editor also reviewed this material before it was published to ensure its accuracy and adherence with BNN Bloomberg editorial policies and standards.
