In early 2026, a joke has been quietly circulating among traders: saying the words “artificial intelligence” during an earnings call is the fastest way to move your stock price. It doesn’t matter what your business actually does. Whether the AI application is real, experimental, or merely a slide deck is irrelevant. Simply utter the words, allow the algorithms to recognize them, and observe the number turn green. In part, it’s a joke. But only in part.
For the better part of three years, markets have been pricing in the AI revolution with an almost religious conviction. They have poured money into chip manufacturers, data center builders, high-profile tech companies, and every software company that promises to incorporate large language models into their product. This zeal has been genuine and, for the most part, warranted. However, during the first few weeks of February 2026, something changed, first slowly and then quickly enough to leave many investors standing at their terminals with puzzled expressions on their faces.
Key Facts: HALO Stocks & the AI Scare Trade
| Concept | HALO — Heavy Assets, Low Obsolescence |
| Coined by | Josh Brown, CEO of Ritholtz Wealth Management & CNBC contributor |
| Popularized | Early 2026, amid a major software stock selloff driven by AI disruption fears |
| Key HALO sectors | Energy & pipelines, utilities, metals & mining, defence & aerospace, transportation infrastructure |
| Notable HALO examples | ExxonMobil, Enbridge, Lockheed Martin, Fortis Inc., Canadian National Railway, Brookfield Renewable Partners |
| AI scare trade trigger | Citrini Research scenario (Feb 2026) projecting AI-driven unemployment above 10% within two years |
| Market impact (Feb 23, 2026) | S&P 500 fell ~1%; IBM dropped 13% — worst single-day loss since 2000; transport sector lost ~$200B in value |
| Brookfield power deal | Signed $10 billion data centre power contract with Microsoft in 2024 |
| Algorhythm Holdings incident | A $15M company with 25 employees caused ~$200B wipeout in transport stocks by announcing AI freight efficiency gains |
| Reference | Roundhill Investments — HALO Stocks Explained |
Almost immediately, the shift was given a name. The phrase “AI scare trade” first surfaced in a research note from Jade Rahmani at Keefe Bruyette & Woods, and by the middle of the month, it was in every financial newsletter thanks to Wall Street’s penchant for labels. What if AI truly works? was the fundamental worry. Not in the gradual, productivity-boosting manner that most projections predicted, but rather in a more disruptive, job-hollowing manner that begins to undermine the economy as a whole.
A scenario note from Citrini Research, published over the weekend and written as though it were from 2028, depicted a world in which aggregate consumer demand had started to decline and unemployment had risen above 10% due to AI-driven job displacement.
The authors took care to refer to it as a scenario rather than a prediction. The fine print didn’t matter to markets. The next Monday, the S&P 500 fell 1%, and IBM, whose decades-old COBOL infrastructure suddenly appeared vulnerable after Anthropic revealed that its Claude Code tool could help modernize that same codebase, fell 13% in a single session, its worst day since 2000.
It’s difficult to ignore how weird everything feels. For years, the market was concerned that AI was developing too slowly, that the returns were not materializing quickly enough, and that the hype was outpacing the reality. AI seems to be developing too quickly these days, which is concerning. Of course, both could be true. However, the whiplash is real and is causing some truly strange behavior, such as a wave of money shifting from software and technology to something that appears to be purposefully dull on the surface.
Josh Brown, the CEO of Ritholtz Wealth Management and a well-known figure on CNBC, came up with the acronym HALO, which stands for Heavy Assets, Low Obsolescence. Brown was searching for a framework to characterize businesses that are outside of AI’s blast radius: companies whose value is ingrained in physical infrastructure that is so large and capital-intensive that no algorithm will be able to replace it anytime soon.
A working oil refinery is unreachable. A model cannot be trained to replace a hydroelectric dam. In a market that has spent years rewarding abstraction, the reasoning is straightforward—almost refreshingly so. Look for items that would hurt if you dropped them on your foot, as one investment writer put it quite nicely. tangible assets. Actual stuff.
In their own way, the industries that fit that description offer a tour of the unglamorous underpinnings of contemporary life. Energy pipelines — Enbridge reached record highs this week, which at the height of the clean-tech frenzy would have seemed almost comical. Due in part to the fact that AI data centers require massive amounts of electricity, utilities and power infrastructure are gaining new attention.
This is because the companies that provide that power are situated at a unique intersection of old-economy reliability and new-economy demand. In 2024, Brookfield Renewable Partners—up 46% from the previous year—signed a $10 billion power deal with Microsoft. Defense and aerospace firms profit from both the increase in military spending during the AI era and the fundamental fact that nations will always require ships, planes, and armored vehicles regardless of what happens in Silicon Valley. Metals and mining, in which there is no software update for the ore underground.
Transportation is more difficult, and it’s worth taking a moment to consider what transpired there in February. Algorhythm Holdings declared that it had utilized AI to boost freight volumes for its clients without adding more employees. Within hours of the announcement, the transportation industry as a whole lost about $200 billion in market value. For background, Algorhythm employed 25 people, had a valuation of roughly $15 million, and sold karaoke machines until August of last year.
On the same day, CH Robinson Worldwide, a freight broker with billions of dollars in revenue and decades of experience, saw a twenty-four percent decline. Algorhythm’s technology might actually be important. It’s also possible that a press release from a micro-cap company that switched from home entertainment hardware to logistics AI in less than a year caused the market, already primed for panic, to overreact. Both are important to remember.
Observing all of this gives the impression that the investing community is attempting to write the story of the AI era before it actually occurs. Although the HALO framework—real assets, durable demand, and physical moats—makes intuitive sense, it is still a wager on the nature of AI disruption, which is still unknown.
The rotation into heavy-asset stocks may be a temporary flight to safety that reverses at the next ChatGPT moment, or it may be a long-term structural change. Markets have a tendency to return to their obsessions. For the time being, investors in pipeline stocks and utility companies are sleeping a little better than those still holding enterprise software because the unglamorous is having its moment.
