print-icon
print-icon
Add ZeroHedge as a preferred source on Google

AI Is Real. So Is The Chop

Portfolio Armor's Photo
by Portfolio Armor
Saturday, Jun 27, 2026 - 13:31

A bull and bear at the England v. Panama game

AI Is Real. So Is The Chop.

This was quite a week. We had tech selloff. Korea contagion. A Micron earnings report that strengthened the AI buildout case. Another Korea crash. More Strait of Hormuz headlines. Oil falling even as shipping risks lingered. And a spotlight on an AI bottleneck trade in one of the most boring component categories imaginable: capacitors.

So how do you trade a week like that? Keep an eye on the big picture, and stick with your process.

That was the point of our trading alerts this week. The AI buildout still has legs, but the market is getting more selective about who benefits from it. That means we want less vague AI adjacency and more exposure to real bottlenecks.

Memory. Optical networking. Timing chips. Power systems. Cooling. Energy storage. Passive components. Data-center infrastructure. Defense electronics. The physical pieces that have to exist before the AI future can scale.

144.82% Versus 10.5%

We started the week with Monday’s alert, built around a simple point: our system has continued to find strong names even in a volatile tape.

The headline number there was from our December 18th Top Names cohort. That cohort returned 144.82% over six months, versus 10.5% for SPY over the same period. 

Of course that kind of return is atypical for a top ten names cohort, and was driven largely by Sandisk's 895% gain over that time frame (the 6-month return for our December 24th Top Names cohort was a less spectacular 25.6%). But it means our system is finding the right sectors: it was in memory names before their explosive run. That's why our current Top Names are always the first place we go to look for alpha. 

Korea Contagion

By Tuesday, Korea had become the center of the AI memory storm.

That was the backdrop for Tuesday’s alert. The selloff in Korea didn’t kill the AI thesis. It did sharpen the question: who captures the economics of the buildout?

The companies buying the infrastructure? The companies selling the scarce components? Or the companies one layer deeper in the supply chain?

That question matters because the market is not treating every AI name the same anymore. Earlier this year, the uncertainty hit software in the “SaaS apocalypse” trade. This week, it hit parts of the Mag 7 complex.

Our response wasn’t to abandon AI. It was to stay focused on bottlenecks and structure.

Recycling Gains

On Wednesday, we wrote about recycling gains.

That is the rhythm we want: exit trades that have worked, then redeploy some of that capital into new setups where the theme, technicals, and option structure line up.

The themes that day were broad, but not random: stored energy and data-center power, edge AI and physical AI semiconductors, a late-stage psychedelic-medicine platform, and domestic drones / defense supply chain.

That mix reflects how we think about the current market.

AI is the dominant theme, but it is not the only one. Reindustrialization, defense, energy infrastructure, and biotech infrastructure can all matter in a world where physical capacity, not just software, becomes scarce.

Micron Passes The Test

Then came Micron.

In Thursday’s alert, we argued that Micron’s earnings were another sign that the AI boom still has legs.

Micron is not a dot-com vaporware story. Its revenue and earnings are moving sharply higher because AI memory demand is real, supply is tight, and the bottlenecks created by the buildout are showing up in the income statement.

That didn’t mean we wanted to chase Micron after the move. It meant we wanted to look for less obvious ways to express the same broader theme.

So Thursday’s trades were tied to AI data-center power and cooling infrastructure, AI compute infrastructure / high-performance computing / memory systems, and space / defense communications.

Same thesis. Different points in the supply chain.

Korea Crashes Again

By Friday, Korea was selling off again.

That was the setup for Friday’s alert: Korea crashes; the AI buildout continues.

Both can be true.

AI is real. It’s a revolutionary technology. Anyone actively using it can see that. And the economic impact will go far beyond chatbots: self-driving vehicles, humanoid robots, other robots, drug discovery, defense, industrial automation, software, logistics, and the replacement or augmentation of skilled human labor.

But the market is still trying to figure out who actually benefits.

That is why we tightened the focus again.

Bottlenecks Over Buzzwords

Friday’s trades centered on three kinds of bottlenecks.

First: precision timing for AI infrastructure. Timing chips are not as glamorous as GPUs, but high-performance computing systems need them.

Second: Multilayer Ceramic Capacitors (MLCCs) and passive components. Zero Hedge had a good piece Thursday on multilayer ceramic capacitors as a potential next AI bottleneck. The basic idea is simple: AI servers don’t just need GPUs and memory. They need power delivered smoothly and instantly to those chips. MLCCs help regulate current, smooth spikes and drops, and filter noise so the system can run reliably.

The market already learned this lesson with memory. A boring component can become explosive when it turns into a bottleneck.

We already had open exposure to one MLCC-related name. Friday, we added more.

Third: biotech infrastructure. Not a binary drug trial. Not a single FDA coin flip. The theme was injectable infrastructure and life-sciences tools: the picks-and-shovels layer behind biologics, GLP-1s, and other injectable drugs.

That is the kind of setup we want now: tangible bottlenecks, and option structures that let volatility work in our favor.

Using Volatility Instead Of Chasing It

The AI thesis didn’t break, but the tape changed. When the tape changes, the filters have to change too.

We put more weight on technical confirmation. We put a little more weight on fundamentals. And we kept using options structures to lower entry cost, harvest volatility, and cap downside.

For example, in one of Friday’s MLCC-related trades, the call we wanted was trading around $13. Instead of simply buying that call, we built a structure around it for about one-tenth of that price.

That is the tradeoff we want: bullish exposure to strong themes, lower entry costs than simply buying calls, and asymmetric upside.

Exits: Eight Winners In The Chop

We had eight full or partial trade exits this week, all winners (our losing options exits tend to collect on OpEx days, because our winners tend to trigger our pre-set exit orders before expiration; you can see examples of those in June’s OpEx day Exits post).

  1. Short calls on Hyliion Holdings (HYLN -8.30%↓). Sold-to-open the October 16th, 2026 $15 calls for $1.03 as part of a 4-leg hybrid combo on 6/11/2026; bought-to-close those calls for $0.20 on 6/26/2026. Profit: 81% on premium collected. Signal: Multibaggers.

  2. Short calls on Lightwave Logic (LWLG -0.51%↓). Sold-to-open the September 18th, 2026 $24 calls for $1.55 as part of a 4-leg hybrid combo on 6/5/2026; bought-to-close those calls for $0.20 on 6/25/2026. Profit: 87% on premium collected. Signal: Market Watchers.

  3. Short call on Urban Outfitters (URBN 2.87%↑). Sold-to-open the June 26th, 2026 $80 call for $2.43 as part of a 4-leg hybrid combo on 5/8/2026; bought-to-close that call for $0.20 on 6/22/2026. Profit: 92% on premium collected. Signal: Chartmill.

  4. Short call on American Superconductor (AMSC 1.31%↑). Sold-to-open the July 17th, 2026 $65 call for $4.71 as part of a 4-leg hybrid combo on 5/26/2026; bought-to-close that call for $0.20 on 6/24/2026. Profit: 96% on premium collected. Signal: Market Watchers.

  5. 4-leg hybrid combo on ACM Research (ACMR 2.40%↑). Entered at a net debit of $3.05 on 5/6/2026; exited the August/June $60/$65 call calendar at a net credit of $7.70 on 6/18/2026; exited the August $45/$40 put spread at a net debit of $0.20 on 6/22/2026. Profit: 146% (return on max risk: 55%). Signal: Market Watchers.

  6. 4-leg combo on Ecopetrol (EC 2.72%↑). Entered at a net debit of $1.40 on 2/20/2026; exited the March $12/$10 put spread at a net debit of $0.05 on 3/12/2026; exited the March/August $14/$13 call calendar at a net credit of $1.60 per contract on 3/19/2026; sold the remaining August 21st, 2026 $13 calls for $4.60 on 6/22/2026. Profit: 339% (return on max risk: 140%). Signal: Market Watchers.

  7. Calls on Nuvectis Pharma (NVCT -3.34%↓). Bought for an average of $2.04 as part of a 3-leg combo on 5/5/2026; sold half of the calls for $9.50 on 6/23/2026. Profit: 366% on the calls sold. Signal: Market Watchers.

  8. Calls on Nuvectis Pharma (NVCT -3.34%↓). Bought for an average of $2.04 as part of a 3-leg combo on 5/5/2026; sold the remaining half of the calls for $17.25 on 6/25/2026. Profit: 746% on the calls sold. Signal: Market Watchers.

The short-call exits were not full trade exits, but they were important. Buying back short calls after most of the premium has decayed removes the cap from the long-call side of those hybrid combos and leaves us with cleaner upside exposure if the stocks bounce back.

We also closed out two full combo trades. One returned 146% on premium, or 55% on max risk. The other returned 339%, or 140% on max risk.

And then there were the Nuvectis Pharma calls.

We sold half of those calls for a 366% gain, then sold the remaining half two days later for a 746% gain.

That is why we try to leave room for runners. Not every runner will do that. But when they do, they can more than make up for a lot of smaller premium-harvesting exits.

Process Over Prediction

Tech weakness. Korea contagion. Hormuz headlines. Oil volatility. Rapid factor rotation. A market trying to decide whether AI spending is a bubble, a bottleneck supercycle, or both.

We don’t need to predict every headline.

We need a process that can handle them.

This week, that process was straightforward: stay bullish where the thesis is intact, tighten the focus where the tape demands it, use option structures to reduce cost and cap downside, harvest premium when it decays, and let the strongest winners become meaningful.

If you want a heads up when we place our next trades, you can subscribe to our trading Substack / occasional email list below.

If you'd like a heads up in real time when we place our next trades, you can subscribe to our trading Substack/occasional email list below. 

Contributor posts published on Zero Hedge do not necessarily represent the views and opinions of Zero Hedge, and are not selected, edited or screened by Zero Hedge editors.
0
Loading...