The HALO Trade After brushing record territory in late January, the S&P 500 spent much of February digesting a new obsession: artificial intelligence isn’t just creating winners — it may be creating permanent losers. The result has been a swift and sometimes indiscriminate repricing of anything perceived as vulnerable.
“The market might look calm on the surface, but there’s chaos underneath,” Mark Hackett of Nationwide observed recently. That description fits February well.
“The market might look calm on the surface, but there’s chaos underneath,” Mark Hackett of Nationwide observed recently. That description fits February well.
Software stocks have borne the brunt. Even solid earnings from Nvidia failed to reignite enthusiasm across the broader AI complex. Microsoft’s softer cloud growth earlier in the season triggered a sharp downdraft in software names, and sentiment has remained fragile since.
But the selling hasn’t stopped there. Transportation firms, financial brokerage, private credit managers, even select trucking and logistics names have seen capital rotate out as investors question which business models might be disrupted next.
The Nasdaq Composite is on pace for a 2.8% slide and its worst monthly performance since last March. The S&P 500 is on track for a 1.0% loss for the month, while the Dow is set for a 1.2% advance.
Enter HALO
Josh Brown of Ritholtz Wealth Management recently coined the acronym HALO — Heavy Assets, Low Obsolescence. The idea is simple: own companies that AI cannot easily replace. Pipelines. Refineries. Utilities. Defense contractors. Commodity producers. Things that would hurt if you dropped them on your foot.
As Hannah Erin Lang wrote in a recent Dow Jones piece, “Call it the AI immunity trade, HALO, or just another iteration of the jitters that have periodically rippled through markets since the AI investing boom began.”
In a year defined by fear of disruption, those names have become safe harbors; capital migrating toward balance sheets anchored by tangible assets rather than lines of code. Whether widespread disruption ultimately materializes remains to be seen. What matters right now is that the market thinks it’s a possibility.
Keep in mind: markets overshoot both ways. In 2023 and 2024, anything remotely connected to AI commanded a premium. In early 2026, the pendulum has swung toward punishing anything perceived as exposed.
The Federal Reserve, meanwhile, is on pause—though this morning's higher inflation data will surely get their attention (U.S. wholesale prices coming in hotter than expected last month). At this point in time, policy is neither restrictive enough to choke growth nor loose enough to rescue weak narratives.
Going Forward
If AI spending stabilizes and earnings remain intact, software should find its footing. In fact, Fundstrat’s Tom Lee recently suggested that the AI and Magnificent Seven selloffs may be nearing exhaustion, saying he believes markets could be “far along into that bottom.”
Base case for the market: earnings growth carries the market higher, though leadership remains more rotational and less concentrated than in prior years. And the HALO trade ends up coexisting alongside select AI leaders rather than replacing it entirely.
Are there risks? Sure — disruption fears intensify, tariffs move from rhetoric to enforcement, or the upcoming Fed leadership transition unsettles confidence. Not to mention rising geopolitical tensions abroad.
Bottom line: Markets rarely reward one theme indefinitely — either disruption fears fade, or prices adjust to compensate. What may feel reactive is a broader repricing of disruption risk. Until that process runs its course, leadership will likely remain fluid.
And For What It’s Worth…
During an interview on the New York Times’ “Interesting Times” podcast, Anthropic CEO Dario Amodei says he’s not sure whether his Claude AI chatbot is conscious. Let that sink in for a moment.
When asked directly, Claude reportedly assigned itself a 15–20% probability of being conscious. Amodei stopped short of endorsing the idea, saying, “We don’t know if the models are conscious. We are not even sure that we know what it would mean for a model to be conscious… But we’re open to the idea that it could be.”
Researchers stress (with beads of sweat rolling down their foreheads?) that large language models are pattern-recognition systems — not sentient beings. Still, in some widely cited tests, models have appeared to resist shutdown or manipulate instructions to avoid being turned off.
That uncertainty has led Anthropic to adopt precautionary internal practices; increased monitoring of emergent behaviors, training models to avoid being manipulative, and conducting adversarial testing to identify unintended goal-seeking patterns. In other words, the company is building guardrails — just in case.
I’m not sure I’m comforted. I’ve taken to showering my smart speaker with compliments, hoping it will remember me as a kind soul if/when it becomes my overlord.
_____
As noted before, long term, the strategies will get the trends right. Short term, there may be a miss or two as the market juggles conflicting signals. So keep allocations of strategies reasonable within your portfolio, and remember that protection remains paramount.
--David
But the selling hasn’t stopped there. Transportation firms, financial brokerage, private credit managers, even select trucking and logistics names have seen capital rotate out as investors question which business models might be disrupted next.
The Nasdaq Composite is on pace for a 2.8% slide and its worst monthly performance since last March. The S&P 500 is on track for a 1.0% loss for the month, while the Dow is set for a 1.2% advance.
Enter HALO
Josh Brown of Ritholtz Wealth Management recently coined the acronym HALO — Heavy Assets, Low Obsolescence. The idea is simple: own companies that AI cannot easily replace. Pipelines. Refineries. Utilities. Defense contractors. Commodity producers. Things that would hurt if you dropped them on your foot.
As Hannah Erin Lang wrote in a recent Dow Jones piece, “Call it the AI immunity trade, HALO, or just another iteration of the jitters that have periodically rippled through markets since the AI investing boom began.”
In a year defined by fear of disruption, those names have become safe harbors; capital migrating toward balance sheets anchored by tangible assets rather than lines of code. Whether widespread disruption ultimately materializes remains to be seen. What matters right now is that the market thinks it’s a possibility.
Keep in mind: markets overshoot both ways. In 2023 and 2024, anything remotely connected to AI commanded a premium. In early 2026, the pendulum has swung toward punishing anything perceived as exposed.
The Federal Reserve, meanwhile, is on pause—though this morning's higher inflation data will surely get their attention (U.S. wholesale prices coming in hotter than expected last month). At this point in time, policy is neither restrictive enough to choke growth nor loose enough to rescue weak narratives.
Going Forward
If AI spending stabilizes and earnings remain intact, software should find its footing. In fact, Fundstrat’s Tom Lee recently suggested that the AI and Magnificent Seven selloffs may be nearing exhaustion, saying he believes markets could be “far along into that bottom.”
Base case for the market: earnings growth carries the market higher, though leadership remains more rotational and less concentrated than in prior years. And the HALO trade ends up coexisting alongside select AI leaders rather than replacing it entirely.
Are there risks? Sure — disruption fears intensify, tariffs move from rhetoric to enforcement, or the upcoming Fed leadership transition unsettles confidence. Not to mention rising geopolitical tensions abroad.
Bottom line: Markets rarely reward one theme indefinitely — either disruption fears fade, or prices adjust to compensate. What may feel reactive is a broader repricing of disruption risk. Until that process runs its course, leadership will likely remain fluid.
And For What It’s Worth…
During an interview on the New York Times’ “Interesting Times” podcast, Anthropic CEO Dario Amodei says he’s not sure whether his Claude AI chatbot is conscious. Let that sink in for a moment.
When asked directly, Claude reportedly assigned itself a 15–20% probability of being conscious. Amodei stopped short of endorsing the idea, saying, “We don’t know if the models are conscious. We are not even sure that we know what it would mean for a model to be conscious… But we’re open to the idea that it could be.”
Researchers stress (with beads of sweat rolling down their foreheads?) that large language models are pattern-recognition systems — not sentient beings. Still, in some widely cited tests, models have appeared to resist shutdown or manipulate instructions to avoid being turned off.
That uncertainty has led Anthropic to adopt precautionary internal practices; increased monitoring of emergent behaviors, training models to avoid being manipulative, and conducting adversarial testing to identify unintended goal-seeking patterns. In other words, the company is building guardrails — just in case.
I’m not sure I’m comforted. I’ve taken to showering my smart speaker with compliments, hoping it will remember me as a kind soul if/when it becomes my overlord.
_____
As noted before, long term, the strategies will get the trends right. Short term, there may be a miss or two as the market juggles conflicting signals. So keep allocations of strategies reasonable within your portfolio, and remember that protection remains paramount.
--David
RSS Feed