Quili AI
85 points by Shugyousha 3 hours ago | 56 comments

vessenes 12 minutes ago
I love the advocacy angle here - it's super well executed. However

> SINCE THE EARLY 2010s, QUILICURA HAS BECOME A MAJOR HUB FOR A.I. DATA CENTERS THAT USE LARGE AMOUNTS OF WATER TO COOL DOWN SERVERS. BECAUSE OF THAT, COUPLED WITH AN ONGOING CLIMATE CRISIS, CHILE IS FACING A SEVERE DROUGHT EXPECTED TO LAST UNTIL 2040.

I do not believe that data centers (Were there AI datacenters in 2010? No) are causing a severe drought in Quilicura that will extend until 2040. I believe climate change could be causing such a drought, and I believe a drought is a terrible thing.

Estimates of water used for cooling are about 0.3mL of water per chatbot query; the 25,000 prompts that this really awesome and fun project answered resulted in 7.2 liters of water "saved" from use as cooling. I would imagine these 40 people drank significantly more than that during the day.

I put "saved" in quotes because no water is destroyed during datacenter cooling. Some (the parts sent to open evaporative coolers) goes back in the atmosphere, and I imagine disperses over a pretty wide region. The rest of it is recycled locally, whether into the water table, or I would guess more likely back into cooling another round of queries.

Anyway, it's fun. But I guess I'd prefer something that is more honest about the actual uses and cycles here -- it doesn't do a local community (or a global one) any good to misinform them about the situation; to their point the situation locally sounds like it could be pretty dire. Ultimately I think it risks falling between performative and exploitative.

reply
jrflo 3 hours ago
Very fun idea. I do think this puts the water argument into a bit more perspective. Your average person should drink ~3L of water per day, or about 22mL per 11 minute average-length chat that Quili quotes on their site. It's hard to pin down exactly how much water AI uses, but something on the order of 5mL per interaction seems reasonable [1]. So the people doing this are using about 4x more water than an AI model would. Of course, it does matter what happens to that water, whether it turns into water vapor or urine and how much of that water re-enters the water cycle, but it's an interesting comparison.

[1] https://www.seangoedecke.com/water-impact-of-ai/

reply
seabass 2 hours ago
The people doing this are drinking water either way.
reply
jrflo 2 hours ago
That's true, but the people could also be doing some other job with their effort too. If you factor in the water consumption of the energy used to power their laptops while answering these questions (~30W of power for 11 minutes per interaction, 1mL of water per W of power) you get another 5mL of water per interaction purely in electricity use.

Not saying it's a 1:1 comparison but it puts it in perspective.

reply
beepbooptheory 2 hours ago
I can't seem to find the laptop power usage bit in the TFA, is that just supposed to be presumed? Isn't that kind of high?
reply
jrflo 2 hours ago
30W is on the low end, based off of web browsing and standard office usage. I didn't link my source so here it is: https://solartechonline.com/blog/how-much-electricity-does-l...
reply
aravpanwar 35 minutes ago
Give them Stillsuits from Dune and we have solved everything.
reply
drcongo 2 hours ago
My urologist told me off for drinking 3L of water per day, said I should be aiming for 3 pints, maybe 4 at the most.
reply
jrflo 2 hours ago
That would translate to 11mL of water per interaction. Plus the 5mL for running the laptop that I pointed out in my other comment. I think the comparison still holds up with that math.
reply
jakeydus 2 hours ago
You sure they weren't just confused about imperial unit conversions?
reply
nater5000 29 minutes ago
This is a "fun" art piece, I guess. But it doesn't make any sense to me. I think part of my confusion comes from this site itself. It really presents this information in a very convoluted and meandering way. Like, it reads like a sales pitch, but it also doesn't provide any of the most interesting information they have available (like the contents of actual chat sessions).

It's also not clear what this aims to prove or highlight. I'm willing to just accept it in the abstract (it's like AI, but with humans), but the whole site is focusing on "saving water" when I don't see how this... event(?) does that in any shape or form. I guess it's just a way to get visibility on the problem? But it does so by pretending like humans are an actual viable alternative to AI for these kinds of chats? Is this supposed to be ironic or tongue-in-cheek? Am I missing the joke?

I don't know. It kind of seems like there was some original intent that got lost along the way, and they tried their best to pull together what they had into some sort of message. But it's lost on me.

reply
beepbooptheory 2 minutes ago
[delayed]
reply
jimmyjazz14 2 hours ago
This is a neat idea but some of the claims on that website seem extremely propagandistic. I'm not particularly gung ho about datacenters but if datacenters are causing water issues the solution is to shine a light on alternative technologies to solve the problem. Perhaps Chile could become a leader in building cleaner datacenters but thats not the goal I am taking from this initiative.
reply
gisely 2 hours ago
This is an alternative technology.
reply
threethirtytwo 31 minutes ago
The problem is ai is better. Thats the haunting part.
reply
biggestlou 2 hours ago
This canard about AI and water usage really just isn't going to go away, is it?
reply
xnx 25 minutes ago
Entirely dependent on the specific datacenter and location. If this datacenter is doing high-volume open loop evaporative cooling in a drought-stricken area, that is pretty bad on the Chilean authorities for allowing it.

The fact that the Panama Canal uses ~2 billion(!) gallons of water a day never gets mentioned.

reply
Synthetic7346 46 minutes ago
> This canard about AI and water usage really just isn't going to go away, is it?

What does canard mean in this context?

reply
neom 2 hours ago
This is cool. It's worth noting that there has been a much wider Chilean water crisis: https://en.wikipedia.org/wiki/Chilean_water_crisis + https://www.bnamericas.com/en/interviews/chile-has-no-time-t...
reply
AnotherGoodName 3 hours ago
Is there a way to read some of the responses that i’m missing? It sounds like a lot of fun to see a what i expect is a Pictionary style human drawn image in response to a prompt. I want to see it.
reply
oneeyedpigeon 2 hours ago
As soon as I saw "Why are hot dog buns sold in packs of 8 and sausages in packs of 12?" I wanted to know the response! My immediate reaction was "because sausages tend to last longer than buns" but I'm curious to see how Quili.ai answered.
reply
ghm2180 3 hours ago
Whoa. This is amazing. I think What is interesting is whats driving this: the psychologically formed habit of asking questions to bots rather than search engines.

If the median query on the internet is now and forever answered by an AI bot, humans could just as easily become really important as an "authentic" source of answers.

reply
toddmorey 2 hours ago
I agree with this, but I can’t help but reflect on how we poisoned our own well by destroying the human internet before AI.

The answers to almost every search are already out there, but we made the public internet so toxic and user hostile, you now need a whole new layer of AI to sift through the bullshit, ads, and clickbait.

I don’t think for a second that AI will not go the same hellish direction.

reply
devmor 2 hours ago
Of course it will go the same direction - many of the leaders in the AI space are the very same entities either technologically or financially that are responsible for that destruction.
reply
devmor 2 hours ago
Unironically this is a "return to tradition" opportunity in the making to me.

If this really is the case, then there's an engineering solution here to re-train people by answering questions with search results first, leaving AI reconciliation for those cases where someone would be clicking the "next page" button or scrolling further.

Of course, that may not align with the perverse incentives driving whoever is behind each bot.

reply
mawadev 2 hours ago
I feel like I'm living in an alternate reality, because I have no idea what this page is about. Something with data centers using up all the water? Reading the comments here and people arguing dcs dont use that much water compared to regular people who drink water, confuses me even more. Like whats next? People are told to drink less and fork over their RAM? That would be wild
reply
BowBun 33 minutes ago
> QUILICURA, CHILE SITS WITHIN THE MAIPO RIVER BASIN, ONE OF THE MAIN SOURCES OF WATER FOR PEOPLE AND FARMING. SINCE THE EARLY 2010s, QUILICURA HAS BECOME A MAJOR HUB FOR A.I. DATA CENTERS THAT USE LARGE AMOUNTS OF WATER TO COOL DOWN SERVERS. BECAUSE OF THAT, COUPLED WITH AN ONGOING CLIMATE CRISIS, CHILE IS FACING A SEVERE DROUGHT EXPECTED TO LAST UNTIL 2040.

> To make the impact visible, and to help draw visibility to the problem, the residents from Quilicura became “human A.I.,” for a day. Answering prompts, creating content with local artists, and completing other tasks manually.

Just read the content my dude. It took me 3 minutes.

reply
tschellenbach 2 hours ago
Sounds like misinformation, let's compare: "The clothing and textile industry uses roughly 170 times more."

The difference is even larger if you look at the state of the art: "As of November 2026, the state of the art is direct-to-chip liquid cooling combined with closed-loop dry coolers. This allows a data center to operate with essentially zero water consumed for cooling, even for high-density AI infrastructure."

You might not want to put an older/less efficient data center in a very dry area, but other than it seems quite reasonable.

reply
etchalon 46 minutes ago
"Clothing uses more!" ... yes. Because humans need clothes.

"Closed loop systems solve this problem!" ... most DCs are not using, or planning to use, a closed loop system.

reply
hridyaagrawal 2 hours ago
Isn't there a bigger water crisis in Chile? datacenters are easy to blame probably
reply
joshstrange 3 hours ago
> Each chat lasted an average of 11’ 4

What does this mean?

reply
ambicapter 2 hours ago
11 minutes 4 seconds
reply
drop_star 2 hours ago
11 foot 4 inches
reply
seabass 2 hours ago
11 minutes, 4 seconds
reply
Forgeties79 2 hours ago
11m4s
reply
josh-wrale 3 hours ago
Humans have neural networks, too!
reply
heckelson 2 hours ago
The beginning of the video made me really thirsty!
reply
idbnstra 23 minutes ago
a thirst trap you might say lolol
reply
rfgplk 2 hours ago
People still don't understand that

a) datacenters don't use water for computation, they use it for cooling

b) the water isn't obliterated, it's still there

c) the water can be recycled and rereleased into the ecosystem, which it mostly is.

reply
idbnstra 23 minutes ago
yeah, even in non closed loop systems, doesn't it just evaporate?
reply
scotty79 2 hours ago
Did they track their water usage for that day?
reply
pelagicAustral 3 hours ago
Who the hell would think putting a datacentre in one of the most actively seismic countries on the planet is a good idea? I'm puzzled. Pretty good initiative other than that.
reply
0x00cl 2 hours ago
Well, usually its not a big problem, Chile in general has a good earthquake resistant infrastructure, because of its building code which has been built based on past experiences.

Chile is a good place, if you look at the submarine cable map (which allows connections over the internet between regions) you'll see that in the west part of South America, Chile is the only country with a submarine cable that goes to Asia, and is going to be working in 2028 and Chile in general over land also has a very good internet connection and is very advanced technologically.

reply
vulcan01 27 minutes ago
People build datacenters in Japan all the time.
reply
gyanchawdhary 2 hours ago
[dead]
reply
eggsbenedict 3 hours ago
"BECAUSE OF THAT, COUPLED WITH AN ONGOING CLIMATE CRISIS, CHILE IS FACING A SEVERE DROUGHT EXPECTED TO LAST UNTIL 2040."

I like the initiative as a life-affirming activity, but spreading falsehoods about water usage is not a good thing. Lying and deceiving makes the world a worse place. This sentence could just as easily read:

""BECAUSE OF [Quili], COUPLED WITH AN ONGOING CLIMATE CRISIS, CHILE IS FACING A SEVERE DROUGHT EXPECTED TO LAST UNTIL 2040."

Without explaining what proportion of the drought comes from the climate crisis, or the mechanism by which data centers cause drought, you could substitute almost anything for [Quili] here.

The water hype needs to go. It's caught on with people who get their news off tiktok, but we need to have better epistemic models if we want to actually make the world a better place.

reply
jmcqk6 56 minutes ago
Why aren't you asking why they are trying to build a datacenter which requires an abundance of water in a place where water is scarce?

Why focus on epistemic models instead of ethical ones?

reply
darksaints 2 hours ago
If you have the same rainfall today as you do tomorrow, but tomorrow someone steals all of the water and leaves you with none, is that a drought? Not technically...but who the fuck cares about the technicality?? You don't have any water!
reply
eggsbenedict 2 hours ago
Yes, in your example, the person who stole the water would have stolen the water.

In real life, in Chile, who "stole all the water" ? How did they "steal" the water?

reply
rhet0rica 2 hours ago
On a related note, many famous famines went from stressful-but-survivable bad harvests to mass casualty events because of mismanagement: the Irish Potato Famine, the Holodomor, the Great Leap Forward...

It seems the villain here could be the Chilean government, for prioritizing the tech economy over the people at a time when it matters most.

reply
etchalon 44 minutes ago
Turns out the real villains were the unlikely friendships between capitalists and government neoliberalism created along the way.
reply
LelouBil 3 hours ago
Really interesting. This makes me wonder if it would cost less to have human answers for like 90% of every-day AI requests.
reply
nater5000 37 minutes ago
>This makes me wonder if it would cost less to have human answers for like 90% of every-day AI requests.

No, not even close lol

reply
djelu 2 hours ago
Excluding the Google AI summaries included with every search, most requests are likely going to be coming from enterprise users using it for code. So, probably no.
reply
noduerme 3 hours ago
This underscores how vastly underrated humans are by the people who are pushing this obnoxious, inhuman, alien technology onto... humans.
reply
wolttam 3 hours ago
I don’t quite get the classification of alien considering they were grown with human data (and humans were the ones that invented the architecture to make it work in the first place)
reply
djelu 2 hours ago
I have a question: have you used one of these models for code, CAD, general 3d work? The vast chasm will take on a very different perspective.
reply
qmmmur 3 hours ago
Totally. This is way more exciting than any new frontier model. It is kind of crazy how good an idea it is...
reply