AI狂热的代价：七千亿豪赌如何引发全球资源短缺

Hello everyone, I'm Zihan. The topic we are discussing today is very explosive and is also closely related to everyone's wallet.

That's right. The Washington Post recently published a deep-dive report revealing a jaw-dropping phenomenon.

This report points out that the frantic spending by tech giants in the field of artificial intelligence is causing serious shortages in all other sectors.

Yes, the article mentioned an astronomical figure: 700 billion US dollars. This is a "cash-burning" war without precedent.

Although this 700 billion dollars in capital expenditure has driven the rapid development of AI technology, it is also like a giant black hole, sucking up the surrounding resources.

What we need to discuss is not just the money issue, but the issue of physical resources. Zihan, can you give everyone a most intuitive example first?

Of course. The most direct impact is—you might not be able to find an electrician.

This sounds a bit incredible. How can high-end AI technology compete for jobs with electricians who fix circuits?

Actually, the logic is simple. The operation of AI relies on massive data centers, and the construction and maintenance of these data centers require extremely complex power systems.

Exactly, like Amazon AWS has built huge data center clusters in places like Boardman, Oregon.

To maintain the operation of these facilities, tech companies are hiring large numbers of skilled electricians and power engineers at high salaries.

This has led to an extreme shortage of electricians in other sectors of the market. Many ordinary construction projects have even been forced to halt because of this.

It's not just labor; the scramble for raw materials has also reached a white-hot stage.

Yes, the report mentioned that orders for some transformers and power generation equipment are already backed up for years.

For those cities in urgent need of infrastructure updates, this is undoubtedly bad news.

Besides infrastructure, this shortage effect has also spread to the consumer electronics sector. I heard smartphones are going to get more expensive too?

That is almost a certainty. The reason lies in chips, especially memory chips.

We know that training AI models requires massive amounts of high-performance memory. This has caused chip manufacturers to adjust their production lines one after another.

Capacity originally used to produce memory for smartphones and PCs is now being largely shifted to produce AI-specific high-bandwidth memory.

This has created a tight supply of ordinary consumer-grade chips, which is typical "supply falling short of demand".

According to predictions, the price of smartphones may continue to rise in the coming years because core components are becoming more expensive.

This is truly "when gods fight, mortals suffer". Before we enjoy the convenience brought by AI, we first have to pay for its costs.

There is another more hidden but far-reaching issue, which is the "crowding-out effect" of capital.

That is exactly what I wanted to say. When 700 billion dollars floods into AI, it becomes very hard for innovative projects in other fields to get financing.

Correct. Many startups in non-AI fields, no matter how good their technology is, are now facing the risk of "running out of food".

Because the eyes of venture capitalists are all drawn to AI; they feel that is the only future.

This lopsided investment frenzy might stifle the sprouts of innovation in other key areas like biotechnology and clean energy.

The Washington Post article even used "unprecedented" to describe the scale of this spending.

This makes one help but think of the dot-com bubble back in the day, but this time it seems to involve the struggle for underlying physical resources even more.

Yes, before they fought for user attention, now they are fighting for electricity, chips, manpower, and capital.

It's hard to say what the outcome of this gamble will be. But the short-term pain seems to be a reality we must face.

For ordinary consumers, perhaps we need to be mentally prepared: phones getting expensive, renovations becoming difficult, and even electricity bills might go up.

But this also reminds policymakers that they need to find a balance point between AI development and guaranteeing people's livelihood.

Very reasonable. Technological progress is certainly important, but it cannot be at the cost of sacrificing the normal operation of society as a whole.

Our time for today's show is just about up. I wonder what views our listeners have on this "resource war" triggered by AI?

Welcome everyone to leave a message in the comments section and tell us your observations. See you in the next episode!

See you!