3 ms·

What the comments fail to point out (and I haven't read the article to find out if they address it): the number of data centers being built is based on speculat

by e40 3mo ago

What the comments fail to point out (and I haven't read the article to find out if they address it): the number of data centers being built is based on speculation of need and is a dramatic over-estimate. It has to be. What, other than AI has happened in the last few years that would warrant that many new data centers? How do they know there will be customers for that capacity? There will not be advances in algorithms, etc that do AI much more efficiently? (We know China is making strides in that!) It's complete speculation! It's the new gold rush and everyone wants a data center. Most of these data centers might not even be built! And the ones that are, might never make any profit.

therealdrag0 3mo ago
It doesn’t “has to be”. You’re speculating just as much.
- e40 3mo ago
  I don't think so. Look at these numbers: | Year | Capacity (GW) | Construction Spending | | 2015 | 0.2 – 0.3 GW | $1.8 Billion | | 2016 | 0.3 – 0.4 GW | $2.2 Billion | | 2017 | 0.4 – 0.5 GW | $3.1 Billion | | 2018 | 0.5 – 0.6 GW | $4.5 Billion | | 2019 | 0.6 – 0.7 GW | $6.2 Billion | | 2020 | 0.8 – 1.0 GW | $8.0 Billion | | 2021 | 1.2 – 1.5 GW | $10.0 Billion | | 2022 | 2.0 – 2.5 GW | $15.0 Billion | | 2023 | 3.5 – 4.0 GW | $20.0 Billion | | 2024 | 6.4 GW | $35.0 Billion | | 2025 | 8.5 GW | $46.0 Billion | | 2026 | 13.6 GW | $50.7 Billion | 2026 figures represent full-year capacity projections and annualized construction spending rates. There is absolutely no way that increase was planned due to precisely planned for capacity. It's speculation.
  - therealdrag0 3mo ago
    What I meant to push back on is that speculation doesn’t mean it’s wrong. It doesn’t “have to be” an “over-estimate” even if it’s obviously speculation. Any multiyear investment that depends on future demand requires speculation. Shrug. I don’t see anything wrong with that. It’s part of running a business and address a market. For all we know it’s an under-estimate.

by e40 3mo ago

therealdrag0 3mo ago

It doesn’t “has to be”. You’re speculating just as much.

e40 3mo ago
I don't think so. Look at these numbers: | Year | Capacity (GW) | Construction Spending | | 2015 | 0.2 – 0.3 GW | $1.8 Billion | | 2016 | 0.3 – 0.4 GW | $2.2 Billion | | 2017 | 0.4 – 0.5 GW | $3.1 Billion | | 2018 | 0.5 – 0.6 GW | $4.5 Billion | | 2019 | 0.6 – 0.7 GW | $6.2 Billion | | 2020 | 0.8 – 1.0 GW | $8.0 Billion | | 2021 | 1.2 – 1.5 GW | $10.0 Billion | | 2022 | 2.0 – 2.5 GW | $15.0 Billion | | 2023 | 3.5 – 4.0 GW | $20.0 Billion | | 2024 | 6.4 GW | $35.0 Billion | | 2025 | 8.5 GW | $46.0 Billion | | 2026 | 13.6 GW | $50.7 Billion | 2026 figures represent full-year capacity projections and annualized construction spending rates. There is absolutely no way that increase was planned due to precisely planned for capacity. It's speculation.
- therealdrag0 3mo ago
  What I meant to push back on is that speculation doesn’t mean it’s wrong. It doesn’t “have to be” an “over-estimate” even if it’s obviously speculation. Any multiyear investment that depends on future demand requires speculation. Shrug. I don’t see anything wrong with that. It’s part of running a business and address a market. For all we know it’s an under-estimate.