To The Moon

📉 AI Hardware Sells Off… But the AI Supercycle Is Far From Over 🚀

The market just gave us a classic example of “sell the news.”

🔻 Nasdaq fell 1.47%

🔻 SOX plunged 4.3%

🔻 TSMC dropped despite reporting an incredible 77% YoY net profit growth

🔻 Memory, storage and semiconductor names were all dragged lower as the VIX jumped nearly 7%.

At first glance, it looks like the AI story is breaking.

I don’t think that’s what’s happening.

Instead, I believe the market is transitioning into Phase 2 of the AI cycle.

Phase 1: Buy Anything Related to AI

Over the past two years, investors rewarded every company connected to AI infrastructure. GPUs, HBM, foundries, networking, cooling, storage—capital flowed aggressively into the entire ecosystem.

Valuations expanded much faster than earnings.

Phase 2: Prove the Returns

Now the market wants evidence.

Hyperscalers have spent hundreds of billions on AI infrastructure. Investors are no longer asking:

“How much are they spending?”

They’re asking:

“When does all this spending generate sustainable revenue and free cash flow?”

That doesn’t mean AI demand is weakening.

It means the market’s expectations have become much higher.

⸻

TSMC’s Results Actually Strengthen the Bull Case

A company doesn’t grow profits 77% YoY if AI demand is falling.

TSMC is the manufacturing backbone of Nvidia, AMD, Apple and many leading AI chip designers.

If demand were truly collapsing:

❌ Utilization would fall.

❌ Margins would compress.

❌ Guidance would disappoint.

Instead, the opposite happened.

Even ASML raised its guidance, reinforcing that advanced-node and EUV demand remains exceptionally strong.

The fundamentals are intact.

The stock simply couldn’t clear an increasingly high bar.

⸻

Why the Entire Semiconductor Sector Fell

This wasn’t company-specific.

This was:

✅ Profit taking after an enormous rally

✅ Higher volatility triggering systematic selling

✅ Portfolio rotation away from crowded AI trades

✅ Investors waiting for hyperscaler earnings to validate AI monetization

The market is repricing expectations—not abandoning AI.

⸻

What Happens Next?

History shows semiconductor stocks rarely move in straight lines.

The strongest bull markets often include sharp 10–20% corrections that shake out late buyers before the next leg higher.

As long as:

✔ Cloud capex remains elevated

✔ AI model complexity continues increasing

✔ HBM demand exceeds supply

✔ Advanced packaging stays constrained

✔ AI inference keeps expanding globally

…the long-term semiconductor thesis remains intact.

⸻

My Take

This feels much more like a healthy reset than the end of the AI cycle.

When companies can report record profits and still see their shares decline, it usually tells us expectations had run ahead of price, not that the business is deteriorating.

The market is becoming more selective—but selectivity is not bearish.

For long-term investors, periods like these often create opportunities to accumulate quality semiconductor leaders at better valuations while short-term sentiment turns fearful.

💎 The AI supercycle isn’t ending. It’s simply becoming more demanding.