Utah Data Center

Good video. I definitely agree the publicly released models are largely marketing hype and have not materialized to the doom and gloom level we read about constantly, yet. However, I have seen how quickly models have improved over the past couple of years, especially when fed by already technical experts in certain arenas. The company I work for just invested around a million bucks in some serious computing power to continue exploring this space. I've seen brilliant engineers train models to help offload some of their work with tremendous success (mostly writing code based on examples of their own code, pattern recognition stuff and data aggregation). These are closed source technically specific models, not the catch-all generic ones most folks are exposed to. We are building tools that help diagnose diseases, and we are using doctors to train and validate them.

It will do most of the heavy lifting for many doctors in the future - review and analyze scans, charts, data, and provide treatment pathways. It will not easily replace the skilled nurse that is more regularly interfacing with the patient. Today, I'd still much prefer my surgery be conducted by a skilled surgeon than a Da Vinci robotic system and the surgeon sitting at a console.

Many big tech companies are laying off a ton of people, and then re hiring them for pennies on the dollar in short term contracts to help train models. They helped develop their replacements internally with the base structure, got laid off, and are working gig work to continue training many other peoples replacements at a more refined level.

I have a friend who owns a successful IT company. He's built lots of automated scripts to run a lot of his regular tasks long before AI. He's now heavily using Claude and loves to talk about it, but I mostly just see him working more trying to sort stuff out to make it function, and haven't seen it pay off for him yet. In reality, he is working for Claude, not the other way around. He was doing this on his own before AI was shoved in our faces, and folks like him will continue to help improve it. He has used Claude for a lot of volunteer work for preserving trail access for MTB around here. Without Claude he would've never invested the time and energy he is now in doing this. It's addictive for him, and many others. We see some benefits as a result, but it is helping these models improve quickly, too. I'm grateful for his efforts here as I want trails to stick around.

My lady is in the web space (Search Engine Optimization) and I even watched her build a pretty robust program (with no programming experience) that can scan and evaluate site performance, provide strategy recommendations, and its results validated by experts in the field to confirm a good model across several companies in completely different spaces. This enables experts like her to cover a lot more ground, and reduces the need for so many people working in that space.

It is certainly coming, and it will have a significant impact on how we live and work. If you don't develop experience and expertise yourself, and give yourself solid foundational knowledge, but instead just use AI to do everything for you, you will be in trouble. This is my concern for younger folks, as they leverage it constantly and don't seem to develop the foundational knowledge and critical thinking skills for themselves first. As Altman said (ironic name) he views AI and brainpower as a utility in the future, something we pay for access to like water or electricity.

AI companies have invested a tremendous amount of money developing their products, and need to show a return at some point. It hasn't quite materialized in the way they pitch it just yet, and the bubble gets bigger. But, I have seen how quickly these things are improving, and that is certainly concerning to me when you zoom out and look at the bigger picture and implications. Not to mention what is happening behind the scenes and the vision some of these AI leaders have, and the power and influence they have, paired with virtually zero oversight or guardrails.

Most important post in here along with T-Mac’s.

As for snatching up 40,000 pristine acres in Utah? FUCK OFF!

How many closed military installations do we have that would suit the need? Damn. We’re not that smart when it comes to accountability. We let greed and stupidity run our world. 

There will come a time when we will have to make a decision on whether to use water for technology or for sustaining life.  

I'm not sure what that decision will be, unfortunately.......

Unfortunately I think we do know the answer. 

less than 20 miles away for me.

just got my sod in last week and it's already drying out because Pineview Water sent all of us a letter saying if we use over the set amount of water per month it's a $500 fine and water gets shutoff. after 2nd offense you don't get secondary water the rest of the year. i won't even be able to get it looking decent this year.

to your point, where is all the fucking water for a project like this going to come from?

We were looking for another water share but the developers have the price close to $30k for an acre foot.   It’s completely insane.  

Are you watering at night…late night? 

I'd be a shame if their water source was somehow impeded miles from the plant.

Old turd breathe spreading his infinite wisdom 🤣🤣🤣

Homey55,
Your reply above might just give birth to a grass-roots resistance movement...
...take away the water, take away the cooling for their fucktarded data collection...
...a little "snipey-snip" here and there to take away the power to run their fucktarded data collection.

Has anyone brought up the local Government in Coweta County in Georgia suggesting the use of eminent domain on family households to support their newest data center they approved? 

I would burn my fucking city down if they tried to eminent domain me out of my estate for a datacenter. 

I didn't say nothin'.

They're trying to build six of the damned things within twenty miles of my house with the closest two being three miles away and upwind of my house, all with natural gas and diesel fired generation. There isn't enough water around here for this and the infrasound is going to be horrific. We bought this place in the middle of a quiet forest reserve two years ago with plans to stay until they dragged us out feet first. Now we're gearing up to ditch the place before property values crash completely. Not sure where to go from here.

What's going to happen after Quantam makes it all obsolete?

My brother is facing the same problem but not with a data center. He’s quite literally in the middle of no where, and a phosphate mining company bought a few thousand acres right across from him. First, they didn’t have permits to touch any of the land near his house. A few months ago they put forth the applications and they had a few town halls on it. Just about anybody who could show up to voice their opinion did, none in favor of the strip mining, but they unanimously approved it as soon as the last citizen sat down. 

About a month after they bought the property, my dad came down and the three of us rented equipment and expanded his small pond and dug it way deeper. No matter how much it rains or anything the damn thing does not fill up at all and I would put everything I own on it being the phosphate mine sucking any water out of the ground. They fuck up everything around here. I was on his porch with him two weekends ago, around 1 am finishing our last beer before I headed back home. All you could hear was machinery working in the dead of night. He’s been talking about selling it before they start mining all the way up to the road. 

It really fucking blows though man. That place was  falling apart when he got it, and him and his wife have done an awesome job turning it into a legitimate Florida cracker style home. It’s where they had their wedding, and still have their Arbor up in the yard that they said their vows under. It will be really hard to say goodbye to that place if they do end up selling. 

The neighbor next door lost his house in the hurricane, and just finished building his brand new 3 bed 2 bath just to think about selling it. 

You guys are riling up some real Hatfield & McCoy feelings about this shit.

Eminent Domain…Shitting all over people that INTENTIONALLY moved AWAY FROM THIS SHIT…

STEALING natural resources. 

Damn. 

Ai is unnecessary and a waste of precious resources. How do these data centers keep getting green lighted but a new mountain bike or dirt bike singletrack takes years of environmental studies for approval. Something is fishy. 

What is an acre foot?

1 acre in footprint 1 foot deep. Basically an American football field covered in 1 foot of water.

AI will die on the vine, because there are only so many Cat Dancing videos the world needs to see. 

All the people pushing this have spent all their money already, and are now cap in hand for funding for something that is doubling in cost when it should be halving, be it land, buildings, IT Infrastructure of natural resources.   And where is the profit coming from, and if it makes any, where is it going ?? 

All the IT Hardware will be out of date in a matter of months ( 24-36) and we are already seeing new 'better' AI making other stuff obsolete before its time. 

Just like EV's . it will developer at a pace that committing at any time , sees you locked into obsolesense from day 1, with no discernable value in what you have bought . 

Maybe you’re right but I don’t see it dying on the vine. It’s going to be used in manufacturing, computing, energy and trading markets.  It won’t be allowed to fail. 

0.98 acres

Those people in my business who are pushing to use AI are those who will find themselves out of work because of it. 

We arent allowed to use memory sticks at all, yet people still think AI will work in our business. 

Big industry protects its IP, and it wont be wanting to feed anything into LLM for AI to use, its just big Google, and big Stolen Google at that.  

My son in law is a big cheese in IT infrastructure and security , and he is already hiring people purely to fix AI related catastrophy's across his business. 

Just like self driving cars, it wont be the ' can we get it to work' that kills it, it will be the ' who is liable when it goes wrong'. 

It wont disappear, but it will be a rush to the bottom just like everything else, and it wont provide the riches that it promised investors at the outset. 

I have heard of Data Centre builds already being put on hold, either through local opposition, or the numbers no longer adding up. 

BitCoin will suffer the same, the cost of 'mining' is growing at a rate that the price doesnt support. 

All I can hope for is, that when the next big thing comes after this ai thing we can convert these abandoned data centers into indoor riding facilities.

When I was a kid, the killdozer/Marvin Heemeyer was a huge meme. The older I get, the more I understand him.