@Binance BiBi Tóm tắt nội dung này
Post
TF_bnb
·
--
🚨 BREAKING : THIS IS HOW DOTCOM BUBBLE REPEATS IN AI IN 2026
The U.S. and Taiwan have just locked in a $500,000,000,000 semiconductor agreement.
$250B from Taiwanese corporations.
$250B from Taiwan’s government.
And U.S. tariffs on Taiwanese goods capped at 15%.
This is being framed as “stability.”
It isn’t.
It is accelerant.
Because the real risk in AI is not chips.
It is profitability.
In 1999, the story was simple: new technology changes everything. Capital flooded in. Infrastructure was built at scale. But profits never caught up to expectations.
The NASDAQ peaked at 5,048 — and then collapsed nearly 77%.
That wasn’t a technology failure.
It was a capital allocation failure.
Now look at AI.
Big Tech is projected to spend roughly $602B in 2026. About 75% of that — nearly $450B — is pure AI infrastructure: GPUs, servers, data centers, and supporting equipment. That’s a 36% jump from 2025.
The spending is real.
The buildout is real.
The profits are not.
Fewer than 40% of companies can clearly link AI to earnings growth. When they do, the contribution is often below 5% of EBIT. According to Morgan Stanley, AI has added only around 0.30% to net margins across the S&P 500 so far.
That single data point explains the entire risk.
We are building the hardware of the future first — just like fiber during the dotcom era — before durable cash flows exist.
Now connect the dots.
A $500B U.S.–Taiwan chip deal.
Tariffs capped to keep supply flowing.
Policy designed to protect and accelerate AI capex.
Translation:
Keep the chips moving.
Keep building AI.
Keep spending — even if profits lag.
This can sustain the narrative longer than most expect. But it also makes the eventual reset sharper.
When spending keeps accelerating and profitability stays thin, the unwind is not gradual. It is violent.
The headlines will celebrate growth.
The balance sheets will tell a different story.
This is how bubbles repeat.
The U.S. and Taiwan have just locked in a $500,000,000,000 semiconductor agreement.
$250B from Taiwanese corporations.
$250B from Taiwan’s government.
And U.S. tariffs on Taiwanese goods capped at 15%.
This is being framed as “stability.”
It isn’t.
It is accelerant.
Because the real risk in AI is not chips.
It is profitability.
In 1999, the story was simple: new technology changes everything. Capital flooded in. Infrastructure was built at scale. But profits never caught up to expectations.
The NASDAQ peaked at 5,048 — and then collapsed nearly 77%.
That wasn’t a technology failure.
It was a capital allocation failure.
Now look at AI.
Big Tech is projected to spend roughly $602B in 2026. About 75% of that — nearly $450B — is pure AI infrastructure: GPUs, servers, data centers, and supporting equipment. That’s a 36% jump from 2025.
The spending is real.
The buildout is real.
The profits are not.
Fewer than 40% of companies can clearly link AI to earnings growth. When they do, the contribution is often below 5% of EBIT. According to Morgan Stanley, AI has added only around 0.30% to net margins across the S&P 500 so far.
That single data point explains the entire risk.
We are building the hardware of the future first — just like fiber during the dotcom era — before durable cash flows exist.
Now connect the dots.
A $500B U.S.–Taiwan chip deal.
Tariffs capped to keep supply flowing.
Policy designed to protect and accelerate AI capex.
Translation:
Keep the chips moving.
Keep building AI.
Keep spending — even if profits lag.
This can sustain the narrative longer than most expect. But it also makes the eventual reset sharper.
When spending keeps accelerating and profitability stays thin, the unwind is not gradual. It is violent.
The headlines will celebrate growth.
The balance sheets will tell a different story.
This is how bubbles repeat.
Disclaimer: Includes third-party opinions. No advice. Binance AI may be used without guarantee. See T&Cs.