Amazon employees ask Seattle to put the brakes on new data centers
theverge.com> “The biggest issue is a belief that AI should be how we solve everything, while ignoring the resources that it costs. This culture is omnipresent across tech.”
Well said. It reminds me of the peak of crypto hype, but worse and more pervasive. There's this attitude that no matter what the problem is, the solution MUST be LLMs.
Datacenters (even the new ones being built today) do a lot more than just AI, though. If the goal is to push back on AI, a moratorium specifically on datacenters rated for more than X megawatts or more than Y acre-feet of daily water per acre of occupied land would've been a much more reasonable approach here than just banning datacenters entirely.
Oh, great ideas. Let's see here:
> The council approved two measures: an ordinance halting applications for data centers with electrical capacity of more than 20 megavolt-amperes — enough power for thousands of homes — and a resolution committing the city to study their impacts as a precursor to permanent regulations. [0]
Well, looks like this is exactly what you're asking for, actually! They have passed a moratorium on datacenters drawing more than 20 megavolt-ampers [1] and have committed to gathering more data before passing permanent regulation.
So, since what the council actually did is exactly what you're asking for here, you support the decision of the Seattle City Council to pass this, right?
[0] https://www.geekwire.com/2026/this-is-seattles-position-on-a...
No, because that's per datacenter, not per datacenter-acre. I'll give them partial credit for trying, but a 5-acre 20MW datacenter and a 50-acre 20MW datacenter are not going to have equivalent use cases or local impacts, and that'll make measuring those impacts more difficult.
Your desired unit of measurement is a stupid and pointless one.
No one measures data centers by “datacenter-acre”, because what matters is resource consumption not power density. A 20 MW is going to consumer 20 WM of power and require 20 MW of cooling, regardless of how many acres it sits on.
Power density is the exact metric by which datacenters are differentiated between “hyperscalar” (i.e. what most people call “AI datacenters”) v. your run-of-the-mill colocation DC.
Also: a flat MW cap per DC is straightforward to game by splitting one big DC into multiple smaller DCs.
The entire process has blind spots. Starting with adding more demand for electricity, more demand for mined materials that harm the people and environment. Amazon hasn't helped the global environmental footprint in any way. If you work in most forms of tech, you've agreed to not lose sleep over these things.
No matter what the problem is, the solution will involve some form of machine learning, inference, or both, with massively-parallel processing that is probably (and unfortunately) centralized to some extent. Hence the need for data centers.
They don't need to be built in the middle of downtown freaking Seattle, though.
> No matter what the problem is, the solution will involve some form of machine learning, inference, or both, with massively-parallel processing that is probably (and unfortunately) centralized to some extent.
I think the point people are making is that this claim is not self-evident, and there's a remarkable lack of justification for it whenever it gets asserted. If you're convinced that this tool can literally solve every problem we have in society, it would help to explain why you're so confident about that. So far all I've heard ever is "exponential growth", which is not particularly convincing when a high school precalculus class gives you enough knowledge to be able to understand that there are curves that look a lot like exponentials before suddenly hitting diminishing returns.
Here in Reno the city council imposed a similar moratorium, buckling under a deluge of NIMBY pressure, citing talking points that are obvious bullshit to anyone who's actually set foot in any of the dozen or so existing local datacenters. All that it accomplishes is a guarantee that future projects and their tax revenue will move to the next county over — and then we'll be wondering why we're still stuck with crumbling infrastructure woefully undersized for our population and an economy dominated by a dying tourism industry.
If I was more tin-foil-hat inclined, I'd hypothesize that this wave of anti-datacenter activism is an astroturfing campaign pushed by the CCP to make sure the US deliberately refuses to compete with China in the technology sector. Or it's an astroturfing campaign by incumbent tech companies to block competition via regulatory capture and grandfathering. Or it's an astroturfing campaign by the agricultural sector (and/or companies like Nestlé) to deflect attention from their multiple-orders-of-magnitude greater water consumption. The reality's probably a lot less exciting, though: just a bunch of people who mean well and are rightfully opposed to Big Tech capitalism, but have been misled (probably by some or all of the above) into throwing out the babies with their bathwater.
Your condescending tone about what people want is crazy. You don’t even acknowledge that this will have a real material impact negative impact on people who are not tech workers. Everyone’s electricity bill will go up. And the argument is always that the benefits will come down stream. But it ignores the real fact that: it will affect people immediately, once electricity bills go up. And all of that for the sake of compute for a technology most people don’t like anyways? Your comment about them being misled implies that they haven’t done their own research to understand the impacts these things have on their communities.
> You don’t even acknowledge that this will have a real material impact negative impact on people who are not tech workers.
Of course I don't acknowledge that, for the same reason I don't acknowledge the “real material negative impact” of wind and solar farms or desalination plants: because there's exceedingly little of it in reality.
> Everyone’s electricity bill will go up.
That's trivial to fix by the electric utility charging datacenters at a higher rate. That's exactly what NV Energy is seeking regulatory approval to do here, and what Seattle's electric utility should be doing (if they ain't already).
>technology most people don’t like anyways?
Most people don't like streaming video and social media? These datacenters ain't exclusively for AI.
>Your comment about them being misled implies that they haven’t done their own research to understand the impacts these things have on their communities.
Correct, and I'll even exply it: the overwhelmingly vast majority of them have indeed demonstrably done zero of their own research on the actual impacts these things will have on their communities and mine. They're entirely parroting talking points fed to them via social media, and those talking points invariably boil down to things that are either not actually applicable to local datacenters or are just outright lunacy on the same level as 5G towers controlling minds via COVID vaccines.
Could be some astroturfing, but compared with fears of drag queen storytellers and wind farms, there is a lot of sentiment ready to be mobilized here. Externalities of data centers on this scale, and the outright lying on what's going on by their builders, are already visible. If there was astroturfing available we'd see the Koch brothers et al ramp up in favor of this, get Tea Party rabble-rousers to insist on data centers. Let's see how China deals with the impact on their land and water too.
I'm glad you're not tinfoil-hat inclined!
Could this headline be rewritten, "Employees of X ask Government to Stop Competition?"
They asked them last week. It's going to vote today.
Build them in Tukwila or Auburn, then.
Why do they need to be built in populated areas at all?
Shorter employee commutes? Shorter last-mile shipping distances? Lower latency to/from local customers? Closer proximity to points of intersection of fiber backbones? Closer proximity to existing electrical/water/sewer infrastructure?
Shorter employee commutes? Shorter last-mile shipping distances? Lower latency to/from local customers? Closer proximity to points of intersection of fiber backbones? Closer proximity to existing electrical/water/sewer infrastructure?
But we're being sold a vision of putting them in low-earth orbit. That means, among other things:
- They don't need to be situated anywhere near their customers
- They don't need a lot of employees to babysit the hardware, or in fact any at all
- They don't need water. Radiative cooling is evidently just fine by itself, even without convection or conduction
- They don't need any networking infrastructure beyond what satellite IP links can provide
- They don't need anything but localized photovoltaic power
Every argument for putting data centers in space applies equally to putting them literally anywhere on Earth.
Orbital datacenters are still hypothetical, at best.
In any case:
>They don't need to be situated anywhere near their customers
They are situated “near” their customers… assuming those customers are also Starlink customers. That's really the only remotely-decent reason (IMO) to put datacenters in orbit: to fulfill the same role for space-based customers (including satellite Internet users) that “edge” datacenters do for geographically-local customers.
Orbital datacenters are still hypothetical, at best.
We're about to find out just how hypothetical they are, since the valuation in the upcoming SpaceX IPO is heavily tied to that particular pipe dream.
Cynically, so that tech companies can get their power usage subsidized by the local population through their higher energy bills.
That's easy enough to fix by charging datacenters at a higher rate than residential customers. Most electrical utility companies already have separate residential/industrial/commercial rates, specifically to prevent large-scale consumers from spiking small-scale consumers' prices.
Here in Nevada, NV Energy's in the process of getting state PUC approval for datacenter-specific “large-load electrical service agreements” specifically to ensure datacenters foot the bill for the infrastructure and generation buildouts needed to support them. Hopefully it goes through, since that seems to me like the exact right way to go about it.