Over $20 trillion in market cap quietly breaking down to new 4-month lows relative to the S&P500.

by






BREAKING: THIS IS HOW DOTCOM BUBBLE REPEATS IN AI IN 2026

The US and Taiwan just locked a semiconductor deal worth $500,000,000,000.

$250B from Taiwanese companies.
$250B from Taiwan’s government.

And the US capped tariffs on Taiwanese goods at 15%.

If they really start doing this to “support US semiconductor operations”…

That is not stability.

That is FUEL for the AI bubble.

Because the AI problem is not “chips”.

It is PROFIT.

In 1999, everyone screamed “new tech changes everything.”
So companies built like crazy.
Then profits did not match the hype.
NASDAQ peaked at 5,048 and later fell about 77%.

That is the dotcom lesson.

Now look at AI.

Big Tech is forecast to spend about $602B in 2026.
About 75% of that, about $450B, is real AI buildout.
Servers. GPUs. Data centers. Equipment.
This is about 36% higher than 2025.

So yes, the spending is REAL.

But here is the simple problem.

Most companies still do not make REAL money from AI.

Less than 40% of firms even link earnings boosts to AI.
And when they do, it is often under 5% of EBIT.
Morgan Stanley says AI has added only about 0.30% to net margins across the S&P 500 so far.

That one statement explains a lot.

We are building the “hardware of the future” first.
Just like dotcom built fiber first.

Now connect the dots.

The US and Taiwan just locked a $500B chip deal.
And the US capped tariffs on Taiwanese goods at 15%.

Translation in simple words:

KEEP THE CHIP FLOW GOING.
KEEP BUILDING AI.
KEEP PUMPING CAPEX.

This can keep pumping even if profits lag.

But the risk is also simple.

If spending keeps exploding while profits stay small,
the reset gets VIOLENT when the mood flips.

I’ve studied macro for 10 years and I called almost every major market top, including the October BTC ATH. Follow and turn notifications on. I’ll post the warning BEFORE it hits the headlines.