Hyperscalers were banned in the Netherlands because compared to their impact they make no money for the country. No jobs, a pitiful amount of taxes yet an enormous need for power, land and water.
That's it. The government will bend backwards to accommodate the port of Rotterdam, Schiphol or ASML because those are national assets. But for AI the math simply ain't mathing.
Sounds smart. There is literally no reasons to rush into AI data centers, it’s pure FOMO. Let the early companies explore the space and the technology mature. If it is here to stay it will still be valuable and you can join the industry.
Being first in something with such a massive cost, with dubious ROI, no good understanding of the hardware lifespan is a very risky bet. It’s completely possible the early players will end up with hardware and infrastructure they have to throw away/upgrade at cost in a few years, just because things are moving so quickly.
It’s fine to have some strategic investment in the grid, infra, and datacenters. But AI DCs in general should be seen as a risky bet
Comments
Hyperscalers were banned in the Netherlands because compared to their impact they make no money for the country. No jobs, a pitiful amount of taxes yet an enormous need for power, land and water.
That's it. The government will bend backwards to accommodate the port of Rotterdam, Schiphol or ASML because those are national assets. But for AI the math simply ain't mathing.
Sounds smart. There is literally no reasons to rush into AI data centers, it’s pure FOMO. Let the early companies explore the space and the technology mature. If it is here to stay it will still be valuable and you can join the industry.
Being first in something with such a massive cost, with dubious ROI, no good understanding of the hardware lifespan is a very risky bet. It’s completely possible the early players will end up with hardware and infrastructure they have to throw away/upgrade at cost in a few years, just because things are moving so quickly.
It’s fine to have some strategic investment in the grid, infra, and datacenters. But AI DCs in general should be seen as a risky bet