As I write this, Bitcoin is coming off of conceivably its worst week ever.
It began out with the January 31, 2026 launch of batch quantity two of the Epstein recordsdata, which implicated none-too-few Bitcoiners and early stage Bitcoin firms (I’m wondering, will we nonetheless be speaking about Epstein in 2036?).
The discharge now reads like a nasty omen. As a result of on Thursday of the identical week, bitcoin suffered its fourth worst drawdown ever, a 21% bludgeoning that bled $16,000 from its worth because it went from $76,000 to $60,000 in a single day.
This was gnarly for bitcoin holders, after all, nevertheless it was gnarlier nonetheless for Bitcoin miners, who have been already struggling beneath traditionally low income compression.
Bitcoin hashprice – a measure of mining income in both USD or BTC per unit of hashrate – hit an all-time low of $28.90/PH/day, in response to Bitcoin mining knowledge platform Hashrate Index. Because of this 1 petahash of hashrate (roughly 5 new technology ASIC miners) would web you a paltry 28 {dollars} and 90 cents.
A bum could make a greater each day wage panhandling.
It’s no shock, then, that Bitcoin’s problem skilled 6 detrimental problem changes (out of seven complete) in three months between November 12, 2025 and February 7, 2026 (and the one optimistic adjustment was 0.04% on Christmas Eve). The final time we had a string of changes like that? 2011.
2011, y’all – when early tinkerers have been mining with the computing energy equal of a toaster in comparison with trendy ASIC miners.
Now, bitcoin’s anemic worth isn’t the one issue weighing on problem. Bitcoin miners are additionally pivoting to AI, and they’re beginning to decommission their ASIC fleets to make room for The Subsequent Massive Factor™.
However the financial stress miners are going through proper now affords an honest glimpse into the way forward for an business whose underlying commodity trades in backwardation on an extended sufficient timeframe. Put one other approach, hashprice is trending to zero, so what does that imply for Bitcoin?
Nothing good. But in addition, nothing dangerous, both.
On the market, blockspace. Used as soon as.
Earlier than we prognosticate, let’s study the place the Bitcoin mining business is now.
I stated earlier that hashprice is trending to zero. This is because of a mix of Moore’s regulation – as semiconductors enhance, so too does the power effectivity of ASIC miners, that means miners can produce extra hashrate with fewer electrons, which places stress on Bitcoin’s problem and reduces the speed of mining rewards per unit of hashrate – and the Halving.
The block subsidy will ultimately hit zero. By 2036, it will likely be 0.78125, so for the block subsidy to supply the identical nominal payout beneath at the moment’s 3.125 BTC subsidy given present BTC costs (roughly $212,000), bitcoin will have to be $272,000.
Failing that, Bitcoin miners higher pray for fats transaction charges. However even right here, the development is working towards them. Proper now, you will get a transaction confirmed for beneath 1 satoshi per digital byte (sat/vbyte).
Bitcoin adoption is at an all-time excessive however the mempool is a ghosttown. A part of that is due to data-efficient upgrades like SegWit and Taproot, nevertheless it’s additionally as a result of Bitcoin is scaling as Hal Finney predicted: through Bitcoin banks, be these exchanges, different custodians, or paper merchandise just like the ETFs.
The one really significant on-chain use of the final three years has come from what some Bitcoiners name shitcoins: ordinals and inscriptions, which mockingly have been largely adopted by “shitcoiners” from the realms of Ethereum and Solana.
Please put aside any moralizing, kvetching, and pearl clutching for a second. It doesn’t matter if you happen to love or hate monkey JPEGs on Bitcoin, however it’s essential to acknowledge that they have been a boon for miners and so they buoyed block rewards earlier than and after the 2024 Halving.
This market is lifeless now, although, and up to now no Layer 2 or various use case for the blockspace has crammed the vacuum they left. Given the shortage of adoption for the swathes of Layer 2 tasks that have been paraded out in the course of the ordinals mania to nice fanfare, I believe it might be smart to not depend on such platforms producing significant charges in 10 years. Hopefully they’ll! However I wouldn’t wager on it.
Possibly within the age of AI folks will begin utilizing Bitcoin timestamps for content material and id attestation – or another, unexpected use of blockspace will pop up – however once more, I’m not holding my breath.
It’s possible, nevertheless, that AI produces no less than some optimistic externalities for Bitcoin miners, even when it additionally brings with it detrimental ones.
The approaching domin(AI)tion blackpill
The most important development in Bitcoin mining during the last yr has nothing to do with Bitcoin.
The biggest Bitcoin miners on the earth – Core Scientific, Riot, IREN, Cipher, CleanSpark, Hut 8, TeraWulf, amongst others – have began swapping ASICs for GPUs to money in on the LLM gravytrain.
It’s nearly a retrograde motion, besides the GPUs aren’t producing nonces for miners as they as soon as did – they’re operating AI or high-performance computing masses.
I’m undecided what number of Bitcoiners have performed this tape by, or contemplated the implications of it. Publicly traded Bitcoin miners – those making these pivots, or no less than those making essentially the most noise about them – account for roughly 40% of Bitcoin’s hashrate. And so they’re looking for a option to convert each foundation level of this complete into computing fodder for Claude, ChatGPT, Gemini, and many others.
Should you’re questioning why, it’s easy {dollars} and cents. They’ll monetize their megawatts for a lot better sums than mining bitcoin. Sorry if that shatters any illusions you’ll have in regards to the fabled altruistic miner who’s hashing to defend the community towards these dastardly dangerous actors.
This can be a good factor truly. Firstly, it’s a headwind for hashrate development, which is a tailwind for mining profitability. Fewer mega miners means extra satoshis to go round for everybody else, however maybe extra importantly, it takes a cohort of Bitcoin miners out of the sport who’ve lopsided operational and financing benefits.
Particularly, I’m speaking about public miners’ entry to capital markets, which permits them to aggressively scale their hashrate even when they aren’t worthwhile. Not making sufficient from mining to cowl your prices? No downside – simply dilute your shareholders! For years, public Bitcoin miners have issued new fairness, offered it into the open market, and used the proceeds to shore up operation prices and increase their operations extra rapidly than personal miners.
The tip result’s that Bitcoin’s hashrate has grown far more rapidly than we’d in any other case count on. When China dominated mining, Bitmain fueled meteoric hashrate development with its self-mining and through the proxies in its spoils system. For the reason that China Mining Ban in 2021 shifted hashrate to the U.S., the speedy proliferation of public miners has had the identical impact.
However the promise of an AI payday will likely be too tempting for these firms to disregard, so this new computing software will take these public miners out of the sport. And this shift will likely be as dramatic as The Nice Hashrate Migration after China’s 2021 Bitcoin mining ban.
The megaminer disintermediation whitepill
This coming change isn’t a blackpill, although. It’s a whitepill.
As mega-miners fade into the background, the smaller and medium-sized miners, those that function on the margins, on the outskirts, and who’ve little likelihood of changing their operations into one other type of knowledge middle, will thrive – or no less than survive.
Ten years from now, the vast majority of hashrate ought to come from these Bitcoin miners, not the publicly traded firms who might mine with out regard for precise profitability. These miners who’re round in 2036 will likely be scrappy, shrewd, and nimble. They may have some edge that makes their operations economical, be that recycling warmth; mining off-grid on oil and fuel wells, wind farms, or photo voltaic arrays; or be built-in on the power-plant stage.
For the few giant scale miners that can nonetheless exist presently, they’ll possible be among the many final bunch in that listing: Bitcoin mining operations that run on energy-producing property, from nuclear websites to pure fuel crops, to absorb extra electrical energy every time there’s a bumper crop of manufacturing.
Maybe it goes with out saying, however after all, this assumes that block rewards are wholesome sufficient to maintain hashrate even on the margins. To return to our math within the second part, bitcoin will have to be no less than $272,000 to match the worth of the present block subsidy.
Ideally, transaction charges make up greater than ~1% of the block subsidy, which has been the theme for greater than a yr, however there’s no assure that this would be the case. (Even when they don’t, although, miners with the bottom value power will nonetheless be mining assuming bitcoin isn’t completely nugatory.)
Power effectivity positive factors from ASICs will assist choose up the slack for general profitability, however solely a lot, because the watt-per-terahash ratio is enhancing at a slower and slower charge and can just about plateau in some unspecified time in the future sooner or later given the present trajectory.
The final 5 years have been the exception, not the rule
However once more, all of it is a good factor, truly, as a result of it is going to disintermediate the biggest actors within the Bitcoin mining business, which consequently function potential chokepoints that would compromise the community.
The general public miners are an apparent centralization level right here. These are extremely scrutinized, legally compliant companies that can bend the knee to Uncle Sam if it threatens their enterprise. (Lest we neglect, MARA (previously, Marathon Digital Holdings) began mining OFAC-compliant blocks – blocks that censored any transaction related to an OFAC-sanctioned Bitcoin pockets – in 2021, even supposing there was no regulation or authorized precedent to mandate such an motion).
Much less apparent, although, is the menace that Bitcoin mining swimming pools current to Bitcoin’s permissionless and censorship-resistant ethos. The overwhelming majority of mining swimming pools function utilizing a full-pay-per-share (FPPS) payout technique. Because of this miners are paid no matter what number of blocks the pool mines, utilizing the hashprice metric we coated within the introduction. This mannequin, the obverse of the pay-per-last-n-share (PPLNS) that Slushpool (now Braiins Pool) pioneered in 2011, signifies that the pool assumes the entire danger of mining, and so they act as insurance coverage firms of kinds for miners by guaranteeing earnings no matter how a lot bitcoin the pool is definitely mining. For instance, if an FPPS pool mines 10 blocks a day and is answerable for 9 blocks value of payouts, they pocket the distinction, but when they mine 8 blocks, they eat the distinction.
As hashprice turns into more and more compressed with every successive block subsidy halving, it is going to turn out to be more and more tough for FPPS suppliers to cowl the chance of mining luck whereas guaranteeing payouts. This turns into much more tough if transaction charges begin making up even a modest quantity of complete mining income, as a result of FPPS swimming pools usually calculate hashprice utilizing the bottom block subsidy plus a rolling common of transaction charges over a given interval. Put one other approach, what occurs when an FPPS pool has to pay its miners utilizing a hashprice that assumes transactions make up 10% of mining revenues, however the blocks this pool mines solely make half of that?
Pool solvency turns into a mounting concern, and so FPPS swimming pools should both adapt, or one other mannequin – both outdated or new – will take its place out of necessity.
That is one other optimistic nonetheless, as a result of it neutralizes one other weak level for Bitcoin. Proper now, Foundry, a U.S.-based mining pool, mines 1/third of Bitcoin blocks. What do you assume would occur if the U.S. authorities tells Foundry to censor sure transactions, and create a white and black listing for accredited or sanctioned Bitcoin wallets?
If FPPS fades into the background, we’d count on self-mining and PPLNS-esque payouts to dominate, and this could eat into the market share of enormous FPPS swimming pools and mitigate the above danger. (The counterfactual to this hypothetical, simply to be intellectually trustworthy, is that as Bitcoin mining turns into extra variable, one or two swimming pools find yourself dominating marketshare, as solely the biggest firms have sufficient sway to draw customers and make good on their payout guarantees).
In the end, Bitcoin mining simply isn’t a very good enterprise, and that’s truly a very good factor. A dwindling block subsidy and hashprice will push mining to the margin, to the bottom value of power attainable, with operators that may solely scale with prudence and diligence. In ten years, Bitcoin mining will possible be far more distributed than it’s now because of this.
It’s totally attainable, then, that we glance again on the mega-mining meta that turned widespread within the U.S. after the China Mining Ban as an aberration fairly than the norm – one other product of a fiat-warped, zero-percent rate of interest coverage financial system that was doomed to run out when the accounting stopped making sense.
Don’t miss your chance to own The 2036 Issue — that includes articles written by many influential figures within the area pondering the challenges of the following decade!
This piece is featured within the newest Print version of Bitcoin Journal, The 2036 Problem. We’re sharing it right here as an early have a look at the concepts explored all through the complete difficulty.
