Concentration of AI stocks inside S&P 500 hits dot-com bubble peak – and Bitcoin miners are now exposed
The 10 largest AI stocks now make up about 41% of the S&P 500, in accordance with a BofA Global Research chart circulated online.
That places the AI basket at the same concentration level that tech and telecom reached across the dot-com peak. The BofA chart put the Nifty Fifty at 40% within the Seventies and Japan at 44% within the late Eighties.
The comparability turns a stock-market focus warning right into a stress take a look at for a nook of crypto that has spent the previous 12 months promoting traders a brand new identification.
The market focus is the stress set off. Miner disclosures and mining reviews provide the publicity map.
Public Bitcoin miners more and more commerce as hybrid infrastructure corporations with BTC publicity. Many have signed AI or high-performance computing contracts, raised capital for denser knowledge facilities, transformed premium energy websites, or shifted investor consideration towards long-term lease economics.
If the AI infrastructure premium fades, these corporations face a distinct type of stress. The threat strikes from hashprice alone into debt, contract sturdiness, building execution, and fairness multiples.
At the identical time, Bitcoin will get a second-order take a look at. A weaker AI buildout might ease the scramble for energy, rack area, interconnections, cooling tools, and GPUs.
That would harm miners whose new valuations depend upon AI development, whereas probably serving to remaining miners if scarce infrastructure turns into simpler to safe.
Miners Have Repriced Themselves Around AI
The miner pivot is now measurable in income forecasts. A projected revenue mix cited by S&P Global Market Intelligence confirmed listed miners together with IREN, Riot Platforms, Core Scientific, HIVE, Cipher, and TeraWulf shifting into AI and HPC workloads.
The projected income combine is already massive sufficient to alter how these corporations are assessed.
Visible Alpha anticipated HPC to account for 71% of 2026 income at IREN and Core Scientific, 70% at TeraWulf, 34% at Cipher, 15% at HIVE, and 13% at Riot.
That unfold exhibits the sector has break up into cohorts. Some miners are changing into data-center operators with Bitcoin publicity.
Others are preserving mining because the core enterprise whereas preserving AI optionality at websites which have energy and grid entry.
The scale exhibits up in miner economics. Public miners have introduced greater than $70 billion in combination AI/HPC contracts, in accordance with CoinShares.
The agency additionally mentioned WULF, Core Scientific, Cipher, and Hut 8 are successfully changing into data-center operators that also mine Bitcoin.
That modifications the market hyperlink from an AI inventory selloff. A falling AI a number of would move via miner equities as a result of traders have assigned worth to the HPC pipeline.
Lower AI demand would additionally stress the financing case for initiatives constructed round long-duration tenants, higher-density cooling, and premium grid positions.
Mining margins would nonetheless depend upon BTC value and problem, however the fairness case would have one other variable.
The leverage knowledge factors in the identical route. CoinShares mentioned a number of miners had taken on massive debt masses for AI buildouts, together with $3.7 billion in convertible notes at IREN, $5.7 billion in whole debt at WULF, and $1.7 billion in senior secured notes issued by Cipher.
CryptoSlate has individually tracked how miners have been funding the AI pivot with debt while selling BTC. Put merely, the AI pivot has added a credit score cycle to a enterprise that already lived with a Bitcoin cycle.
The desk mixes 2026 income estimates, 2025 firm disclosures, and contract updates, so every row alerts publicity throughout completely different time horizons.
| Miner | AI/HPC publicity sign | Repricing stress level |
|---|---|---|
| Core Scientific | Visible Alpha projected 71% HPC income share in 2026 | CoreWeave supply, customer-funded capex, conversion execution |
| TeraWulf | 522 important IT MW underneath long-term leases | Financing, tenant timelines, credit-enhanced contract supply |
| IREN | AI cloud ARR goal above $500 million from 23,000 GPUs | GPU contract period, utilization, tools economics |
| Riot | 600 MW Corsicana AI/HPC analysis | Value of utilizing premium energy for AI versus mining |
| Cipher | Visible Alpha projected 34% HPC income share in 2026 | Debt-funded HPC buildout and website monetization |
Cipher’s rebrand toward HPC provides one other instance of the shift. TeraWulf’s Fluidstack expansion exhibits how miners have paired massive energy portfolios with AI tenants and credit score assist.
The Risk Is In The Sites, Contracts, And Capital Stack
Core Scientific is the cleanest instance of the shift from mining sensitivity to infrastructure execution. In its fourth-quarter 2025 results, the corporate mentioned it had energized about 350 MW underneath its CoreWeave contract and remained on observe to ship about 590 MW by early 2027.
It additionally reported that This fall colocation income rose to $31.3 million from $8.5 million a 12 months earlier, whereas digital asset self-mining income fell to $42.2 million from $79.9 million.
That is the pivot in working type. Power and buildings as soon as tied primarily to Bitcoin manufacturing are being monetized via colocation.
Core Scientific additionally mentioned $226.2 million of its $279.2 million in fourth-quarter capital expenditures was funded by CoreWeave underneath current agreements. That buyer funding reduces some capital pressure, however it additionally exhibits how deeply the buildout relies on an AI tenant’s development path.
The conversion has accounting complexity as properly. Core Scientific mentioned it was restating prior monetary statements after figuring out improper capitalization of property dedicated to demolition throughout facility conversion from mining to HPC colocation infrastructure.
The subject was company-specific, however it illustrates a broader level. Moving from mining halls to high-density AI infrastructure goes past advertising language.
Core Scientific’s canceled CoreWeave merger agreement exhibits that AI-linked worth already sits inside shareholder selections.
CoreWeave’s 2025 Form 10-K provides counterparty context, together with massive contracted energy commitments and disclosed dangers tied to AI demand.
The miner publicity is due to this fact linked to each website supply and the monetary well being of the AI cloud ecosystem.
TeraWulf exhibits the identical shift at a bigger contracted scale. In its full-year 2025 results, the corporate reported long-term knowledge heart lease agreements totaling 522 important IT MW, greater than $12.8 billion in long-term credit-enhanced buyer contracts, and $6.5 billion in long-term financings.
It mentioned HPC internet hosting had grow to be its major development engine whereas it continued to function legacy mining infrastructure opportunistically.
CoinShares reported that WULF mined 262 BTC in This fall alongside $9.7 million in HPC lease income. The similar report mentioned WULF’s cost-per-BTC figures have been distorted by the corporate’s transition, together with curiosity, SG&A, and depreciation linked to the brand new infrastructure base.
That distinction is essential. Once a miner turns into an AI infrastructure firm, per-BTC price metrics can distort the enterprise until the steadiness sheet is separated from the remaining mining fleet.
Riot’s Corsicana choice exhibits how AI optionality can alter Bitcoin’s capability path earlier than a ultimate AI contract even exists. The firm’s Corsicana update mentioned it was evaluating AI/HPC makes use of for about 600 MW of remaining energy capability, halting a beforehand introduced 600 MW Phase II Bitcoin mining growth, and slicing anticipated year-end 2025 self-mining capability from 46.7 EH/s to 38.4 EH/s.
IREN provides a distinct publicity kind. Its October 2025 AI cloud update focused greater than $500 million in annualized AI cloud income from 23,000 GPUs by the tip of Q1 2026, with 11,000 GPUs contracted for about $225 million of ARR on common two-year phrases.
That creates a quicker repricing channel than long-term colocation. GPU cloud economics can shift as {hardware} provide, utilization, and buyer budgets change.
Power Scarcity Sets Bitcoin’s Side Of The Trade
The Bitcoin aspect of the commerce is much less direct. A weaker AI infrastructure cycle would stress miners with AI publicity first via fairness worth, funding price, and contract expectations.
Bitcoin’s community would really feel the change via the economic base that competes for a similar energy and websites.
The AI-mining link is bodily. Bitcoin mining stays the bigger combination income pool in key BTC value situations, whereas AI has grow to be a right away financial threat to the community’s industrial safety base.
AI and mining compete for land, grid interconnections, substations, cooling design, financing, and administration consideration.
Energy demand from AI explains why that competitors is sturdy. The IEA estimated knowledge facilities consumed about 415 TWh of electrical energy in 2024 and projected world data-center consumption would roughly double to 945 TWh by 2030 in its base case.
AI-driven accelerated servers account for a significant share of the rise. Data facilities could be constructed quicker than energy programs can add transmission, substations, and era, which makes location and grid entry beneficial.
A North America data-center trends report provides the market bottleneck behind that argument. Low emptiness and high preleasing make power-ready capability extra beneficial.
For miners, the scarce asset is commonly the energized website, with the ASIC fleet just one half of the stack.
As of press time, Bitcoin market data exhibits BTC buying and selling close to $76,800, with a market cap round $1.5 trillion, a present block reward of 3.125 BTC, and community hashrate above 1.1 ZH/s.
CryptoSlate’s aggregate market page exhibits Bitcoin dominance round 60% of a $2.6 trillion crypto market. Those figures put miner economics underneath stress even earlier than AI competitors is taken into account.
BTC value, charges, problem, and power price nonetheless resolve how a lot safety spend Bitcoin can assist.
A cooling AI cycle might ease one half of that stress. If hyperscaler demand, GPU shortage, or data-center preleasing weaken, miners that stayed nearer to Bitcoin might discover energy websites and infrastructure much less contested.
Difficulty might regulate if capability exits mining, elevating hashprice for remaining operators. That mechanism seems in CryptoSlate’s analysis of miners as AI utilities.
That aid has limits. The fee and cost picture retains the upside certified, with charges close to zero and price stress close to $80,000 per BTC.
Difficulty aid alone leaves weak miner economics unresolved. Long-term AI leases, customer-funded buildouts, interconnection agreements, tools specialization, and website conversion prices additionally create lag.
An AI unwind would launch capability inconsistently, and some of it could stay unusable for mining at engaging returns.
Two Outcomes Depend On AI Demand
The market threat signaled by the AI focus chart results in two completely different outcomes for miners.
In the primary, AI demand holds. Public miners with high-quality energy campuses preserve signing HPC contracts as a result of AI tenants can provide longer income visibility than Bitcoin mining.
Premium websites preserve drifting towards AI, whereas mining concentrates round versatile energy, demand response, stranded power, and geographies the place interruption is suitable.
In that state of affairs, public miner equities grow to be much less dependable proxies for BTC as a result of enterprise worth comes from leases and data-center execution as a lot as mined Bitcoin.
In the second, AI infrastructure reprices. The miners most exposed to AI development face stress via leverage, fairness multiples, contract assumptions, and building pipelines.
Debt raised for data-center buildouts turns into tougher to hold if anticipated AI returns fall. GPU cloud contracts with shorter phrases can reset quicker.
Long-term colocation leases might provide extra safety, though in addition they lock websites right into a path which will take years to reverse.
Bitcoin’s potential profit sits downstream from that injury. The upside is a loosening of scarce inputs, decrease competitors for energy, and a greater hashprice atmosphere for operators nonetheless targeted on mining.
It is an industrial-security argument, with BTC value sitting outdoors the direct declare.
That is why the AI focus chart belongs in a Bitcoin-miner balance-sheet dialogue. The chart raises the likelihood that the AI commerce has grow to be crowded.
The miner knowledge exhibits which crypto corporations have constructed round that commerce. The unresolved take a look at is whether or not these AI/HPC contracts stay sturdy sufficient to justify the shift, or whether or not the identical infrastructure that pulled public miners away from Bitcoin turns into a supply of stress.
For Bitcoin, the outcome can be combined as a substitute of clear. A repricing might weaken some of the best-capitalized public miners whereas making power and data-center inputs much less scarce for the miners that stay.
The subsequent sign will come much less from AI rhetoric than from financing phrases, tenant supply schedules, new energy contracts, and hashprice. Those are the variables that can present whether or not miners purchased a stronger enterprise mannequin or imported a second cycle into Bitcoin’s safety base.
The publish Concentration of AI stocks inside S&P 500 hits dot-com bubble peak – and Bitcoin miners are now exposed appeared first on CryptoSlate.
