Why Bitcoiners turned to AI
The 2024 Bitcoin halving lowered block rewards to three.125 BTC, slicing miners’ earnings in half. This transformation, mixed with increased electrical energy prices, costly gear upkeep and elevated competitors, made conventional mining much less worthwhile. Many mining corporations struggled to keep up their revenue margins and commenced exploring different income sources.
Though Bitcoin mining depends on gadgets known as ASICs, mining corporations have entry to energy-dense knowledge facilities and energy infrastructure. As demand for AI compute skyrockets, many miners are repurposing or upgrading their services with GPUs to help AI coaching and inference workloads.
Nonetheless, synthetic intelligence calls for immense computing energy, particularly for coaching massive language fashions, powering autonomous methods and operating enterprise AI instruments.
As tech corporations race to safe high-performance infrastructure, Bitcoin mining corporations are stepping in. Leveraging their present energy-intensive knowledge facilities and upgrading with GPUs, many miners have begun providing AI cloud companies or renting out spare capability. This diversification permits them to generate regular, non-crypto earnings streams, lowering reliance on risky Bitcoin (BTC) revenues.
This shift offsets the impression of Bitcoin halving and has led to extra worthwhile and steady income streams.
Do you know? AI workloads and Bitcoin mining each demand large power. By planning for each, miners can lease extra capability to AI corporations, particularly throughout crypto downturns, turning stranded energy right into a steady money circulate.
Case research: Core Scientific’s $3.5 billion lifeline
Core Scientific is a powerful instance of how shifting to AI might help a struggling Bitcoin mining firm get better. After going through monetary difficulties and submitting for Chapter 11 chapter in late 2022 as a result of low Bitcoin costs and heavy debt, the corporate restructured and returned to the Nasdaq in early 2024.
In June 2024, Core Scientific signed a 12-year, $3.5 billion contract with CoreWeave, an AI cloud computing firm. The settlement allowed Core Scientific to make use of components of its infrastructure to help CoreWeave’s high-performance computing wants, shifting away from solely mining Bitcoin to additionally offering AI companies.
Though the corporate’s income within the first quarter of 2025 fell to $79.5 million from $179.3 million the earlier yr, the AI technique boosted investor confidence. The corporate’s inventory value rose after the CoreWeave deal was introduced, reflecting market help for its new route.
By mid-2025, CoreWeave restarted talks to accumulate Core Scientific, following an unsuccessful $1 billion provide the yr earlier than. This renewed curiosity highlights how the corporate’s deal with AI cushioned the impression of Bitcoin’s halving and positioned it as a key participant within the rising AI computing business.
Hut 8’s AI aspect hustle
Hut 8 has added AI as a secondary supply of earnings whereas persevering with to prioritize Bitcoin mining. This enterprise mannequin combines stability and development potential via a five-year contract that features mounted funds and a revenue-sharing element, guaranteeing regular earnings with alternatives for added earnings based mostly on buyer success.
In September 2024, the corporate launched Highrise AI, a subsidiary providing GPU-as-a-Service utilizing over 1,000 Nvidia H100 chips, specialised {hardware} for coaching and operating superior AI fashions. This transfer marked Hut 8’s official entry into the high-performance computing (HPC) market.
Regardless of its AI enterprise, Hut 8 stays devoted to Bitcoin mining. Within the first quarter of 2025, it mined 167 BTC, a lower from 716 BTC in the identical interval of 2024, largely as a result of 2024 Bitcoin halving. The corporate continues to put money into its mining infrastructure, supported by its important Bitcoin reserve of 10,273 BTC, making it the ninth-largest company Bitcoin holder worldwide.
For Hut 8, AI serves as a complementary technique, diversifying its income whereas protecting Bitcoin mining because the core of its long-term plan.
How are hybrid fashions gaining traction: Hive and Iren
As Bitcoin mining income shrink, hybrid fashions combining mining with AI compute are gaining floor. Firms like Hive and Iren are proving that it’s doable to develop AI income with out abandoning their Bitcoin roots. They’re diversifying earnings whereas optimizing present infrastructure.
Hive Digital Applied sciences
Previously often known as Hive Blockchain, the corporate rebranded in mid-2023 to mirror its broader high-performance computing ambitions. Hive invested $30 million to deploy Nvidia-powered GPU clusters, marking a decisive pivot towards AI workloads.
This funding started to repay rapidly. In fiscal 2025, Hive’s AI and HPC internet hosting income tripled to $10.1 million, virtually 9% of its whole income. Wanting forward, Hive has set an bold goal of $100 million in AI income by 2026, signaling a powerful dedication to increasing its hybrid mannequin.
Iren (Iris Vitality)
Australian mining agency Iren started its AI journey in early 2024 with simply 248 GPUs, and by mid-2025, it had scaled as much as greater than 4,300 models. The agency’s hybrid mannequin is already producing outcomes, mining 1,514 BTC in Q3 FY2025 whereas pulling in $3.6 million from AI cloud companies. To help this development, Iren is constructing AI-focused knowledge facilities in Texas and British Columbia.
Nonetheless, the corporate faces a problem: A category-action lawsuit filed in October 2024 alleges it misled buyers in regards to the operational readiness of its Texas facility, casting a shadow over its in any other case promising growth.
How main Bitcoin miners are getting ready for AI: Riot Platforms and MARA Holdings
Whereas some Bitcoin miners have already begun incomes income from AI, others are constructing foundations for future AI alternatives. Riot Platforms and MARA Holdings, two main corporations within the mining business, are strategically planning for AI integration whereas sustaining their deal with Bitcoin mining.
Riot Platforms
Exploring AI potentialities, Riot Platforms has began assessing the potential to transform 600 megawatts at its Corsicana, Texas, facility into high-performance computing (HPC) infrastructure. Though Riot has not but secured important AI contracts, its Corsicana web site, protecting 355 acres, has the capability to help as much as 1 gigawatt of computing energy, giving it a decisive benefit.
Financially, Riot stays sturdy in its major enterprise, having mined 1,530 BTC and earned $142.9 million in mining income within the first quarter of 2025. The corporate additionally holds 19,225 BTC (as of July 17, 2025), one of many largest company Bitcoin reserves worldwide.
MARA Holdings
MARA possesses essentially the most in depth Bitcoin treasury amongst mining corporations, with 50,000 BTC, second solely to Technique amongst public corporations. Its AI technique focuses on edge computing, together with creating its MARA 2PIC700 immersion cooling system, designed to deal with intensive computing duties.
Whereas MARA has the infrastructure prepared, its AI efforts haven’t but resulted in important contracts or constant income. For now, a transfer into AI stays a forward-looking technique with potential for future development.
Do you know? Bitcoin mining depends on ASICs, however AI wants GPUs like Nvidia’s H100s. Some miners at the moment are retrofitting knowledge facilities with GPUs to help AI purchasers, creating dual-purpose infrastructure that balances each blockchain and AI calls for.
An outlier case: Canaan’s retreat from AI
Whereas many Bitcoin mining corporations are exploring AI to broaden their earnings sources, Canaan has taken a unique method.
In July 2025, the corporate closed its AI chip division, stepping away from the high-performance computing sector. This resolution displays a renewed deal with its major experience: designing application-specific built-in circuits (ASICs) for Bitcoin mining.
As a substitute of pursuing the rising AI market, Canaan is advancing its mining {hardware} to keep up a aggressive edge. Nonetheless, it holds solely 2.1% of the worldwide ASIC market, far behind main rivals like Bitmain and MicroBT.
By prioritizing mining-focused {hardware} and strengthening its presence in markets like North America, Canaan is adopting a singular technique when others are shifting towards AI. The long-term success of this method is but to be decided.
Do you know? AI corporations face strain to go inexperienced. Bitcoin miners that already use renewable power, like hydro or photo voltaic, can entice AI purchasers seeking to meet sustainability targets via clear colocation offers.
Key dangers and issues for miners getting into the AI market
As Bitcoin miners more and more shift to AI, this transition gives alternatives and important dangers. Miners should rigorously think about the next:
- Infrastructure prices vs returns: Shifting from ASIC-based mining to GPU-based AI methods requires substantial preliminary funding. Miners should make sure that the potential long-term income outweighs these prices.
- Consumer stability: AI purchasers, notably startups, could lack constant funding or long-term reliability. Miners ought to rigorously consider purchasers to keep away from fee defaults or service interruptions.
- Energy provide reliability: AI operations demand steady, high-energy utilization. Miners should safe steady, long-term energy agreements and monitor native grid capability to stop outages or sudden value will increase.
- Cooling and thermal administration: AI chips, resembling Nvidia H100s, produce important warmth. Insufficient cooling methods can result in gear failures or lowered effectivity.
- Regulatory compliance: Internet hosting AI workloads could contain complicated rules associated to knowledge privateness, mental property, worldwide knowledge internet hosting, power use, water consumption and carbon emissions. Miners have to be ready to navigate these guidelines.
- Market competitors: As extra miners enter the AI colocation market, pricing may decline. Early entrants ought to set up benefits, resembling strategic areas, low power prices or large-scale operations.
- Useful resource pressure: Increasing into AI whereas sustaining mining operations could overstretch monetary and administration assets.
