Lots of places still pushing forward against people’s wishes though. Construction is real easy to interrupt though and these over leveraged companies can’t afford the delays. It’d be real easy to derail this shit if people start acting.
You can't tell whether it's an Ai datacenter or one that merely holds and protects your health data and DMV records. Or a mix.
The criteria needs to not be the PURPOSE of a DC, since that's impossible to gauge before or after the build, but the build of the DC.
no water cooling
ya pay for your power
no fucking jet engines outside
noise bylaws strictly enforced
You know, like everyone else.
One of my contracts is with a gov adjacent org. I know they've got your personal info - the stuff like your DMV recs - kept very safely in a datacenter somewhere. It's safe, reliable, and has great data connections.
Tear these down and ... what? Back to paper in an office somewhere with no backup?
I do find value in LLMs as a tool for specific uses, but no way I’m paying through the nose for an expensive subscription to support greedy, shithead tech bros plundering the world for profit and government agencies using the tech to spy on everyone.
Open weight models running on consumer hardware are getting pretty good, though. After learning some tricks from the Codacus YouTube channel, I’m running the Qwen 3.6 MoE model on my 4060 gaming laptop with 200k context and usable performance for about the same electricity cost as playing a game. Open source agents like OpenCode are also getting really good. I’m using the beta desktop version of OpenCode and it’s pretty close in capability and UX to the Claude agent I am forced to use at work.
There are still some rough edges and performance issues, but the local LLM scene is already quite decent and is rapidly evolving. Qwen 3.7 open weight models are supposed to be released soon and I’m also looking for things like TurboQuant, speculative prefill (PFlash), etc. to mature and smooth the rough edges.
Anyone outsourcing critical thinking to LLMs is using them wrong. With software development, for example, quality is speed. Yes, a codebase riddled with technical debt will always collapse under its own weight, regardless of whether humans or LLMs wrote the shitty code, so using LLMs to generate slop that senior engineers aren’t carefully reviewing, or letting LLMs make architectural decisions is the wrong approach. However, skillful use of LLMs can actually yield better quality code than humans alone. Here are some examples:
LLMs often find issues when reviewing PRs that the human reviewers missed in my experience. Different, focused agents can be used to review code from different perspectives, like security, performance, etc.
Test coverage can be higher with LLMs automating the test writing. They can often also come up with useful test cases that humans didn’t consider and also make it possible to maintain more varied types of tests that the team otherwise might not take on.
LLMs are great at fixing lint issues. I’ve seen codebases where lint warnings accumulate because only the errors are getting fixed, but LLMs can fix everything quickly.
Once solid patterns have been established in a codebase by senior engineers and documented in agent MD files, LLMs can implement features and follow the existing patterns for routine features and bug fixes without getting “creative” and introducing new tools and patterns that complicate things out of boredom.
LLMs can be used to look at codebases for open source dependencies to better understand how they work in order to utilize them better
LLMs can suggest alternative libraries, approaches, patterns, etc. with the pros and cons of each. While humans still need to do their own research, this can often be a useful starting point that helps make more informed decisions
You can use Claude code with your own LLM too. It seems to be better at automatically creating memories and stuff. Downside is of course that it's closed source.
Codex is also open source and usable with your own LLMS. Same for Mistral Vibe but that's been much less useful for me, opencode with their free or Go plan models is much better. Opencode Go with Deepseek Flash gets you so much usage it's hard to run ouy at 10 bucks a month
19 Comments
melsaskca@lemmy.ca · 35 pts · 103d
Good. Put the 130 B towards food, rent, and healthcare for the people.
areakode@riskeratspizza.com · 11 pts · 103d
ThAt'S cOmMuNiSm!
melsaskca@lemmy.ca · 6 pts · 102d
Gasp! What have I done!? /s
usernametbd@lemmy.zip · 17 pts · 103d
Lots of places still pushing forward against people’s wishes though. Construction is real easy to interrupt though and these over leveraged companies can’t afford the delays. It’d be real easy to derail this shit if people start acting.
Mosfar@sh.itjust.works · 11 pts · 103d
Good!
MalReynolds@slrpnk.net · 8 pts · 103d
So, what's that as a percentage? Of successful actually running 'data centers'
Seems a lot like they actually want to be stopped because it's not profitable, it's the idea that's profitable.
Mwa@thelemmy.club · 8 pts · 103d
Good riddance if its a AI datacenter
corsicanguppy@lemmy.ca · 2 pts · 102d
You can't tell whether it's an Ai datacenter or one that merely holds and protects your health data and DMV records. Or a mix.
The criteria needs to not be the PURPOSE of a DC, since that's impossible to gauge before or after the build, but the build of the DC.
You know, like everyone else.
One of my contracts is with a gov adjacent org. I know they've got your personal info - the stuff like your DMV recs - kept very safely in a datacenter somewhere. It's safe, reliable, and has great data connections.
Tear these down and ... what? Back to paper in an office somewhere with no backup?
Knock_Knock_Lemmy_In@lemmy.world · 4 pts · 102d
How about number of GPUs as a metric. The more, the worse the datacenter?
phutatorius@lemmy.zip · 8 pts · 103d
That's a good start.
EatMyPixelDust@lemmy.blahaj.zone · 6 pts · 102d
Good, we don't need AI corporations destroying the environment just to keep the bubble lasting a little bit longer
funkless_eck@sh.itjust.works · 3 pts · 102d
Georgia Power put out an ad recently claiming data centers help the economy.
Motherfucker read the room. Everyone's aware the "economy" has no bearing on the actual lived experience of the economy.
jaykrown@lemmy.world · 2 pts · 103d
Good, we shouldn't be building any new data centers at all.
goatinspace@feddit.org · 1 pts · 103d
130billion is 29 Kung Fury 2 movies
Whostosay@sh.itjust.works · 3 pts · 102d
What an astoundingly better investment
melfie@lemmy.zip · 0 pts · 102d
I do find value in LLMs as a tool for specific uses, but no way I’m paying through the nose for an expensive subscription to support greedy, shithead tech bros plundering the world for profit and government agencies using the tech to spy on everyone.
Open weight models running on consumer hardware are getting pretty good, though. After learning some tricks from the Codacus YouTube channel, I’m running the Qwen 3.6 MoE model on my 4060 gaming laptop with 200k context and usable performance for about the same electricity cost as playing a game. Open source agents like OpenCode are also getting really good. I’m using the beta desktop version of OpenCode and it’s pretty close in capability and UX to the Claude agent I am forced to use at work.
There are still some rough edges and performance issues, but the local LLM scene is already quite decent and is rapidly evolving. Qwen 3.7 open weight models are supposed to be released soon and I’m also looking for things like TurboQuant, speculative prefill (PFlash), etc. to mature and smooth the rough edges.
prole@lemmy.blahaj.zone · 6 pts · 102d
Have you considered not outsourcing critical thought to a glorified autocorrect?
I dunno, just a thought.
You're part of the problem.
melfie@lemmy.zip · 1 pts · 102d
Anyone outsourcing critical thinking to LLMs is using them wrong. With software development, for example, quality is speed. Yes, a codebase riddled with technical debt will always collapse under its own weight, regardless of whether humans or LLMs wrote the shitty code, so using LLMs to generate slop that senior engineers aren’t carefully reviewing, or letting LLMs make architectural decisions is the wrong approach. However, skillful use of LLMs can actually yield better quality code than humans alone. Here are some examples:
boonhet@sopuli.xyz · 1 pts · 102d
You can use Claude code with your own LLM too. It seems to be better at automatically creating memories and stuff. Downside is of course that it's closed source.
Codex is also open source and usable with your own LLMS. Same for Mistral Vibe but that's been much less useful for me, opencode with their free or Go plan models is much better. Opencode Go with Deepseek Flash gets you so much usage it's hard to run ouy at 10 bucks a month