Meta Inks a New Geothermal Energy Deal To Support AI (theverge.com)
- Reference: 0178034491
- News link: https://hardware.slashdot.org/story/25/06/13/0318228/meta-inks-a-new-geothermal-energy-deal-to-support-ai
- Source link: https://www.theverge.com/news/686387/meta-ai-data-center-geothermal-energy-xgs
> Geothermal plants generate electricity using Earth's heat; typically drawing up hot fluids or steam from natural reservoirs to turn turbines. That tactic is limited by natural geography, however, and the US gets around half a percent of its electricity from geothermal sources. Startups including XGS are trying to change that by making geothermal energy more accessible. Last year, Meta made a [2]separate 150MW deal with Sage Geosystems to develop new geothermal power plants. Sage is developing technologies to harness energy from hot, dry rock formations by drilling and pumping water underground, essentially creating artificial reservoirs. Google has its own partnership with another startup called Fervo developing similar technology.
>
> XGS Energy is also seeking to exploit geothermal energy from dry rock resources. It tries to set itself apart by reusing water in a closed-loop process designed to prevent water from escaping into cracks in the rock. The water it uses to take advantage of underground heat circulates inside a steel casing. Conserving water is especially crucial in a drought-prone state like New Mexico, where Meta is expanding its Los Lunas data center. Meta declined to say how much it's spending on this deal with XGS Energy. The initiative will roll out in two phases with a goal of being operational by 2030.
[1] https://www.theverge.com/news/686387/meta-ai-data-center-geothermal-energy-xgs
[2] https://hardware.slashdot.org/story/24/08/30/007231/meta-strikes-geothermal-energy-deal-to-power-us-data-centers
This is called "Hedging your bets". (Score:2)
Meta is just contracting everything that can in theory provide electricity. They would contract an ICE prison full of alleged illegal aliens on home trainers, if such thing exists.
It does not mean that all of this has a chance to be a success.
Re: (Score:2)
While true, there is a decent chance this could work. There is a lot of promising new tech in this area, ironically some of it coming from the fracking industry. Geothermal works well, the issue has always been the limited number of places you can install it. A lot like extracting gas, it wasn't practical to access some of it until people figured out how to frack relatively cheaply.
Re: (Score:3)
The problems are manifold. Basel, Switzerland has had a geothermal project. They had to shut it down because it caused too many earth quakes. Basically, the insurance for all the damage was far larger than the revenue from the energy they provided.
Re: (Score:2)
Are they doing fracking in Switzerland? That has to be at least as damaging as a geothermal project, and it happens everywhere there are oil deposits in the US. Apparently we are willing to accept a few earthquakes in exchange for energy.
On the plus side (Score:2)
If AI fails, that's more renewable always-available energy.
LLMs are stupid (Score:2)
The only advances is in pointless wasting of electricity.
I saw this coming, it's about the lowest cost. (Score:1)
When looking at varied studies on the cost of different energy sources a trend should appear to the reader: [1]https://en.wikipedia.org/wiki/... [wikipedia.org]
The lowest cost options, excluding fossil fuels. tends to be onshore wind, hydro, nuclear fission, and geothermal. Companies that want low cost energy and some good PR for being "green" will choose one of those four. How is it that there was so much investment in solar power and offshore wind previously if it was so expensive? Because some people convinced the fede
[1] https://en.wikipedia.org/wiki/Cost_of_electricity_by_source#Global_studies
Yet another reason why "AI" is a failure (Score:1, Insightful)
If this stuff ever takes off for real and becomes a daily part of everyone's lives the way these tech bros hope, where is all the power coming from?
These are very early days and they already need new power plants just to support this software in its infantry. How much more energy and thus new power plants will be needed if it goes mainstream?
The grid in the US, EU, and many other places is already fragile as we've seen many times over the years. Good luck with the new fully loaded power requirements.
Re: (Score:1)
'software in its infantry'.
An army of Agent Smiths?
Re: Yet another reason why "AI" is a failure (Score:4, Insightful)
Human species was already investing more in tech than agriculture globally before "AI". Add electric cars and evermore air conditioning to the grid too. It doesn't fail enough already. We are not an intelligent species.
Re: (Score:1)
This reminds me of a phrase that I recall was attributed to Winston Churchill, it goes something like how Americans can be trusted to do the right thing but only after all other options have been exhausted. Another way I've heard this behavior on seeking the "right" solution over the best solution, only to have the "right" solution fail, is that reality can be denied for only so long. There's been denying reality on a need for "firm" electricity production for a long time, people trying the "right" soluti
Re: (Score:2)
It seems that with supply increases, price rises would only occur due to market manipulation. We may be headed towards some kind of Russo-Saudi-US-(Israeli? weird, I know) oil alliance that intentionally keeps other supplies offline until whatever runs the USA now can further dominate those countries (Iraq, Venezuela).
Re: (Score:2)
Indeed. And they obviously do not expect energy needs to get much lower as these are strategic energy investmenst. But here is a real problem: That energy costs money. Will the rather pathetic business models ever be able to generate enough revenue to pay that electricity bill? I have some serious doubts on that front alone. And then there are all the other problems with the current AI hype.