I’ve been wanting to chat about this for a bit, seems like there’s a lot going on with data centers lately, and I think it’s a good time to bring it up.
So, is running a data center really more profitable now than mining Bitcoin? This isn’t just a meme anymore. A bunch of big Bitcoin miners are ditching mining to jump into the data center space. You don’t really see anyone discussing it, but it’s happening pretty fast behind the scenes.
I noticed some chatter on Twitter, aka X, where people are saying AI data centers are raking in more cash. What do you all think? If AI data centers start pulling in more profits, what’s gonna happen to Bitcoin mining in the long run?
AI is more of a bubble than BTC is. Which is not a bubble at all.
And I do think it's much more in needed funds to reward ratio than for mining, depending on what scale we speak of
It is worth knowing that this has been talked about on some threads.
The big miners that have gone for AI data centres do not completely leave bitcoin mining, they are still mining bitcoin.
About bitcoin, there is no problem, it can not get to a stage all the miners will leave for AI data centres. Bitcoin mining also remain profitable.
If bitcoin become not really profitable, some miners will leave, making it profitable again for miners that do not stop mining.
AI data centers are currently a threat to bitcoin mining profitability due to competition for power infrastructure and driving up energy costs that lead mining companies to turn into AI infrastructure providers instead.
Currently, AI workloads can offer profits 25x higher per kilowatt-hour compared to bitcoin mining. And its predicted that by the end of 2026, 70% income of the mining companies will be obtained from AI and high-performance computing (HPC) contracts.
However, this will also serve the ''lifeline'' for mining companies to stay in profits through offering rentals of their pre-built infrastructures and power access to AI.
I think they're just more stable, not more numerous. That's if they can successfully compete with established data center companies like Google and Amazon.
The important question is, how long will this bubble last?
Mining will remain profitable as long as it offers block rewards and fees; not all pool companies will have the same interest in the data center business. Especially with fewer competitors, the opportunity to dominate the hashrate is greater.
I do not think AI data centers are taking over Bitcoin mining so much as they are competing for the same cheap power, land, and financing. That part is real. If a miner can get materially better returns by pivoting part of the site into HPC or leasing power to AI tenants, some of them obviously will.
Companies follow margins, true, but that does not mean mining disappears. Mining has one big advantage a lot of people keep glossing over: it is far more flexible. ASICs can sit next to ugly power sources, they can curtail fast, and they do not need the same pristine uptime and cooling profile that AI workloads want.
You missed The first miner has already run towards AI and few other threads.
I wouldn't worry about it too much. If enough miner stopped mining and switched to different business, mining difficulty would be reduced so other miner would have more profit. Besides, miner also must consider what to do with ASIC they no longer use and transition cost.
I know very little, to nothing, about BTC mining and profitability. From what I understand, the AI data center craze is having a very big come-to-jesus moment. Also a little bit of eyes being too big for the stomach situation going on. BTC mining industry seems more set-in-stone and carry less risks despite profitability being volatile from a short-term perspective. We are undoubtedly in an AI bubble, and the recent Anthropic source-code leak confirmed it if you read between the lines (literally) of what was found and what the implications are in terms of assesing what exactly investors have been incinerating throwing their money at.
If I had the money, means, and motion to build, operate, and periodically upgrade/maintain 40 data centers or 40 BTC mining farms, I'd choose the latter. The ever-increasing hashrate over the fully zoomed out timeframe is more promising than "planned" and "stalled".
This is a very interesting topic. AI data centers are more profitable business than Bitcoin mining and they also offer much better long-term stable contracts and growing institutional backing. But here is the thing, Bitcoin's price is very low right now. If it reaches $100K and higher, Bitcoin mining might recover and AI pivot might slow down. Honestly, I'm shocked that Bitcoin worth so low right now.
AI is a bubble at some point but it's really useful. It can significantly help you to analyze large data, improve the work efficiency and make the whole working process more productive. Let's be honest, AI brings lots of benefits on the table. If you have only used ChatGPT, then you might think that it's trash but Claude is really another level.
It's important to consider the short-term vs long-term impact of both before deciding which is more profitable.
AI data centers may look more profitable at the moment but that alone doesnt mean mining should be pushed behind. Mining has always gone through cycles and survived which shows that it has a built in adjustment mechanism. When profits drop, some miners leave, difficulty adjusts and things balance again.
Just like @BattlDoge pointed out, miners can pivot if returns make more sense elsewhere so its not really a fixed path.
The real pressure point here is competition for cheap power not AI itself, that's what I think.
A data sender seems to give a more stable income than Bitcoin mining. It is not out of context for these miners to diversify into other profitable sectors. Some AI companies are avoiding the delay caused by construction, so they offer good deals to Bitcoin miners, which some of them cannot refuse.
I think some may still switch back to mining during the bull run when it would become profitable. There will always be Bitcoin miners to sustain the ecosystem, so there is no cause for alarm.
Source for this claim?
Nothing guarantees that an increase in price doesn't mean an increase in hasing pwoer bring bought back online and thus decreasing the revenue earned per th/s, considering how many have pulled out of the mining game, that would mean a ton of cheap gear is now available and ready to be picked or just monthballed and would be ready in weeks to be bought back online.
It also had a ton of disadvantages
- A crash in bitcoin price leaves you with gear that has absolutely no usage whatsoever
- Cheap power on the other side of the world can put you out of business
- A new and more advanced ASIC will always outperform you
Also, the part about cooling and uptime is misleading, cooling is the result of power used, a 2MW datacenter will need the same cooling as a 2MW mining farm.
This iss what I could see online about it:
About the second quote, the website to the analysts referenced is not found if you visit the site, but it is very clear that AI is competing with bitcoin, although just like I have explained above, not a problem.
The miners will also have to think about it because bitcoin miners can prefer to shut down their miners or some of their miners but AI miners are working 24/7. I am not sure about this but what I saw on the first article which you can see the link on the first quote.
It is worth knowing also that some miners may decide not to sell their mined coins until the price of bitcoin increased significantly. Another thing is the mining difficulty which can reduce and miners make more profit.
I have friends on social media in the mid-west that are constantly complaining about data centers. They say their water is polluted and no longer safe to drink. They say their food has been tainted and isn't the same. They complain about prices rising and disrespect to locals...
I am even seeing that they just broke ground on a new massive data center about 15 miles from my house. People are extremely angry that their pristine flower farms are being turned into massive ugly buildings.
Pretty crazy situation. This AI bubble needs to pop before we build the Matrix.
There is a saying " A sacred place is never empty", which means that others will always be willing to take the place of those who leave a particular field of activity. If some people start focusing on AI data centers instead of bitcoin mining, new participants will come to take their place in mining. Moreover, these two activities can be done in parallel, which is likely what those who previously focused only on bitcoin mining will start doing.
This isn't a problem at all, because when miners switch to the AI sector, it means there will still be some left, and with fewer competitors, the remaining miners will still earn more.
Although the AI sector currently offers more lucrative profits than Bitcoin mining, not all miners are willing to switch to it because they would need to spend more resources and considerable money to adapt to AI's needs.
Therefore, there's nothing to worry about because miners will still be around since Bitcoin mining continues to generate profits for them.
I like the advancement of the Digital space ongoing of which has led to AI innovation and technology being incorporated and invested in with a better assurance of profit as compared to Bitcoin.
I agree also that Bitcoin mining won't die but would get to balance itself out for those left on the network with less competition and better pay, just as AI assures to provide.
Although the energy cost is what's making AI better than Bitcoin mining, it won't affect the market but create a new niche that it appeals to on the global market.
Sure a lot of interest from computational data centers is shifting from Bitcoin to AI, but you just need to understand that this is not a problem for Bitcoin. Bitcoin mining is a self-correcting market. When hash power leaves, that just makes it more profitable for miners, which will bring in more opportunities for miners to expand or new miners to join the network. It stays in equilibrium. That's one of the many brilliant things about Bitcoin that makes it unstoppable.
Bitcoin has had "global level" network security for many years now and the hash rate has gone up many times over during those years. So security is not at threat even if a bunch of miners switch partially or fully to AI training.
Also realize that AI training is already very troubling for the amount of money being put into it and the amount of energy it is consuming. It's likely a bubble and the money being put into absolutely dwarfs the money that was ever put into Bitcoin mining. Bitcoin is fine and there is absolutely zero threat for miners or network security no matter what happens with all this AI data center stuff.
the asic gear can run in dirty hot areas better than servers.
so I would argue it is easy to do a one or two container mine. vs a one to two megawatt server setup.
I read all this and you need to think of 2036 we are under 180,000 coins in 2036.
a block is 1.56 in 2028
a bloc is 0.78 in 2032
a block is 0.39 in 2036
a block is .19 in 2040.
so if i am a 20 megawatt guy with twenty containers mining can i dish off 2 to 4 meagwatts for ai as a hedge.
i know I would want to hedge into ai.
here an other reason to hedge ai.
which a big company mara has done.
mara has a mine in texas.
300 mega watts more than 300 1 megawatt containers.
they have a clever power deal with the grid. they will turn gear off when texas grid is over loaded and texas grid pays them not to mine.
so they keep 240 containers and lease power to ai .
they do not turn the ai off only the miners .
this is cheaper then upgrading from s21xps at 13 watts to hydro a 9.5 watts.
so they certainly were going to do 15 to 20% switch to ai.
this means no expense to make more money. as they will let ai company build a site and lease 50-60 megawatts to them.
this eases the burden of changing over 300 containers of single phase s21xps to three phase hydro s23s
but mara is a giant.
look at diff about 155t in november is now around 136t over 10% drop in hash.
but price dropped a lot since the fall.
Are you looking from a business owner's perspective? Or do you find many use cases for AI in your workload other than asking LLM to summarize some info for you? I do think LLM/AI is helpful for me to some extent, not the point of relying on it 24/7, though. My best use case for them is to help me learn languages and some coding. I'm not sure I'd pay online services for that and I don't think I need datacenters to do what my GPU can do, although it's likely manufacturers would make it very expensive for enthusiast go get a hand on a new GPU in the next years.