Replies

Here’s my personal ( albeit imperfect) take on your questions.

The massive new data centers are being built because AI workloads, cloud computing, and hyperscale storage require orders of magnitude more compute than the older internet ever did.

Yes, they use very large amounts of electricity, and many (not all) use significant water for cooling—those concerns are real, but vary by design and region.

And yes, they create very few permanent jobs relative to their size and cost—this is also true.

our Privacy Statement.
Today
Please answer these questions for me as objectively as you can.

The internet and cell phones as we know them have been around for decades, and have functioned without all these massive data centers that are apparently being built. So why is there such a need for the data centers?

And what are the actual facts? Opponents say they use up too much electricity, and too much water. Is that true or not? We hear conflicting things about that.

What about jobs? Some people say these data centers create jobs. Other people say these data centers create virtually no jobs in a community.

What is the reality?
Here’s the clear, objective reality, broken into the three issues you asked about: why the new data centers exist, electricity and water use, and jobs. I’ll give you the concise answer first, then the deeper breakdown.
Concise Takeaway
The massive new data centers are being built because AI workloads, cloud computing, and hyperscale storage require orders of magnitude more compute than the older internet ever did.
Yes, they use very large amounts of electricity, and many (not all) use significant water for cooling—those concerns are real, but vary by design and region.
And yes, they create very few permanent jobs relative to their size and cost—this is also true.

1. Why do we suddenly “need” huge data centers?

Because the internet of the 2000s and smartphones of the 2010s did not require the computational intensity that AI, cloud computing, video streaming, and global-scale storage require today.

What changed?

Here are the major drivers:

AI model training and inference. There has been a HUGE DEMAND for the productive capabilities of AI

Training frontier models requires thousands of GPUs ( microprocessors ) running for weeks.

Serving AI queries (inference) at global scale requires enormous ongoing compute.

Then there’s Cloud migration

Corporations moved their entire IT infrastructure into the cloud. Instead of thousands of small server rooms, workloads consolidated into hyperscale centers.

Then there’s Video streaming…

Netflix, YouTube, TikTok, Twitch—video is the dominant internet traffic category.

Because of these, we now have Data storage explosion. Every company stores logs, analytics, backups, and customer data in the cloud. Global data volume doubles roughly every 2–3 years.

Bottom line:

The old internet didn’t need hyperscale compute.
AI and cloud computing do.

2. As a result of these demands, the need for MORE ELECTRICITY has accelerated!

A single hyperscale data center can consume 50–100 megawatts.
Some reach 300–500 MW (equivalent to a small city).

AI-focused centers can exceed 1 gigawatt of planned capacity.

Why so much?

GPUs and CPUs generate heat; cooling them requires massive power. AI workloads are especially energy-intensive. These then require another thing — COOLING!

Many data centers use evaporative cooling, which consumes millions of gallons of water per day in hot climates.

In places like Arizona, Oregon, and parts of Europe, water use has become a political issue.

3. Jobs: do data centers help local employment?

Here’s how I see it AS IT APPLIES TO DATA CENTERS ONLY. The flow on products and other demand as a result of what they provide is an entirely different matter. But just to focus on these data center build outs alone…

Data centers create many construction jobs, but very few permanent jobs.

During Construction phase, about 1,000–3,000 temporary jobs over 1–3 years are created to build the data centers. Local contractors benefit. This is the main economic boost.

After construction, comes the Permanent operations.

Here, a hyperscale data center typically employs about 50-100 full-time workers, Mostly technicians, security, and facilities engineers, I’ve even heard of some centers operating with fewer than 30.

Why so few? Because:

Servers are automated

Cooling systems are automated

Monitoring is centralized

Maintenance is periodic, not continuous

So, Opponents are correct that data centers do not create large numbers of long-term jobs. Supporters are correct in a sense that they create high-paying technical jobs, but the number is small.

But that’s not the entire story, we have to look at the BIG PICTURE — Here’s the objective reality: data centers create very few direct jobs during their build out and operations, but they do create broader economic effects — just not in the way politicians often claim. The key is understanding where the job creation actually happens, and where it does not.

The key to understanding the “bigger picture” is this: data centers don’t produce physical goods — they produce CAPABILITY and PRODUCTIVITY. And that capability is what drives job creation outside the data center itself and the economy at large— manufacturing, finance, entertainment, education, information technology, etc.