Why Bitcoiners turned to AI
2024 bitcoin halling reduced block rewards to 3.125 BTC, which cuts the income of miners in half. This change, high power cost, expensive equipment combined with maintenance and increased competition, made traditional mining less profitable. Many mining companies struggled to maintain their profit margin and began searching for other revenue sources.
Although bitcoin mining depends on devices called Asics, mining companies have access to energy-erosion data centers and power infrastructure. As a demand for AI’s Compute Skycket, many miners are re -presented or upgraded with GPU to support AI training and estimates.
However, artificial intelligence demands immense computing power, especially to train the large language model, to strengthen autonomous systems and run the enterprise AI equipment.
As a race for tech companies to secure the infrastructure of high performance, bitcoin mining firms are being stepped into firms. To take advantage of their existing energy-intensive data centers and upgrade with the GPU, many miners have started offering AI cloud services or have started renting extra capacity. This diversification allows them to generate stable, non-crypto income currents, reduces the dependence on volatile bitcoin (BTC) revenue.
This shift turns off the effect of bitcoin and has produced more profitable and stable revenue currents.
Do you know Both AI workload and bitcoin mining demand large -scale energy. By planning for both, miners can lease extra capacity to AI firms, especially during crypto recession, transform stuck power into a stable cash flow.
Case study: $ 3.5 billion lifeline of core scientific
Core Scientific is a strong example of how shifting for AI can help fix a struggling bitcoin mining company. After facing and filing financial difficulties for chapter 11 bankruptcy in the end of 2022 due to low bitcoin prices and heavy loans, the company restructured and returned to the NASDAQ in early 2024.
In June 2024, Core Scientific signed a contract of $ 3.5 billion with AI Cloud Computing Company Korwave. The agreement allowed the core scientist to use parts of its infrastructure, to support the high-demonstration computing needs of the Korwave, which provides AI services away from mining bitcoins.
Although the company’s revenue fell from $ 179.3 million to $ 79.5 million in the first quarter of 2025, the AI strategy promoted investor’s confidence. The company’s share price increased after the announcement of the Coreweave deal, which reflects the market support for its new direction.
By mid -2025, the corevoy resumed the negotiations to acquire the core scientific, one year ago after offering a failed $ 1 billion. This renewed interest highlights how the company’s meditation on AI gave the impact of bitcoin’s urge and deployed it as a prominent player in the growing AI computing industry.
AI side hustle of Hut 8
Hat 8 has added AI as a secondary source of income to prioritize bitcoin mining. This business model combines stability and growth capacity through a five-year contract consisting of fixed payment and a revenue-sharing component, which ensures stable income with additional income opportunities based on the success of the customer.
In September 2024, the company launched Hirise AI, a subsidiary of GPU-e-A-A-Service using more than 1,000 NVidia H100 chips, special hardware for training and special hardware for upgraded AI models. The move marked the official entry of Hat 8 in the high-demonstration computing (HPC) market.
Despite its AI enterprise, Hut 8 bitcoin remains dedicated to mining. In the first quarter of 2025, it mined 167 BTC, a decrease from 716 BTC in the same period of 2024, roughly 2024 bitcoins. The company continues to invest in its mining infrastructure, supported by its important bitcoin reserve of 10,273 BTC, which makes it the ninth largest corporate bitcoin holder worldwide.
For Hut 8, AI serves as a supplemental strategy, keeping bitcoin mining as the origin of its long -term plan, brings variety in its revenue.
How hybrid models are receiving traction: Hive and Ireen
As the bitcoin mining profits shrink, the hybrid models of a combination of mining with AI calculations are receiving the ground. Companies like Hive and Ireen are proving that it is possible to increase AI revenue without leaving their bitcoin roots. They are bringing diversity in income by optimizing the existing infrastructure.
Hive digital technologies
East Hive Blockchain was known as the company Rebrand To reflect its widespread high-performance computing ambitions in mid-2023. Hive invested $ 30 million to deploy Nvidia-operated GPU groups, marking a decisive axis towards AI workload.
This investment started early paying. In FY 2025, the AI and HPC of Hive hosted revenue, which increased to about 9%of the total revenue, $ 10.1 million. Further, Hive has set an ambitious target of $ 100 million in AI revenue by 2026, indicating a strong commitment to expand its hybrid models.
Ireen (Iris Energy)
Australian mining firm Ireen began its AI trip with just 248 GPUs in early 2024, and by mid -2025, it had increased to over 4,300 units. The hybrid model of the firm is already producing results, drawing 1,514 BTC in Q3 FY2025, drawing from AI cloud services to $ 3.6 million. To support this development, Irene is building AI-centric data centers in Texas and British Columbia.
Nevertheless, the company faces a challenge: a class-action case filed in October 2024 has been accused of misleading investors about the operational readiness of his Texas facility, otherwise put a shadow on promising expansion.
How the major bitcoin miners are preparing for AI: riot platforms and mara holdings
While some bitcoin miners have already started earning revenue from AI, others are building foundations for future AI opportunities. Two major companies in the mining industry are strategically planning for AI integration, focusing their focus on riot platforms and Mara Holdings, bitcoin mining.
Riot platform
AI possibilities discovered, riot platforms have started assessment The ability to convert 600 MW in its Corsicana, Texas, facility in high performance computing (HPC) infrastructure. Although the riot has not yet secured important AI contracts, its course site, covering 355 acres of land, has the ability to support up to 1 GW of computing power, giving it a decisive benefit.
Economically, the riot remains strong in its primary business, mined 1,530 BTC and earned $ 142.9 million in mining revenue in the first quarter of 2025. The company has 19,225 BTC (till 17 July, 2025), one of the largest corporate bitcoin reserves worldwide.
Mara holdings
Mara has the most broad bitcoin treasury among mining companies, with 50,000 BTC, second after a strategy among public companies. Its AI strategy focuses on edge computing, including its Mara 2PIP700 immersion cooling system, which is designed to handle intensive computing tasks.
While the infrastructure is prepared near Mara, its AI efforts have not yet resulted in significant contracts or frequent revenue. For now, a step in AI has a forward -looking strategy with potential for future development.
Do you know Bitcoin mining depends on Asics, but AI requires GPU like the H100s of NVIDIA. Some miners are now rebuilding data centers with GPU to support AI customers, creating a double -objective infrastructure that balances the demands of both blockchain and AI.
An external case: Canaan Retreat from AI
While many bitcoin mining companies are searching for AI to broaden their income sources, Canaan has taken a different view.
In July 2025, the company shut down its AI chip division, stepping away from the high -performance computing sector. This decision focuses a renewed focus on its primary expertise: designing the application-specific integrated circuit for bitcoin mining.
Instead of carrying forward the growing AI market, Canaan is carrying forward its mining hardware to maintain a competitive lead. Nevertheless, it is only 2.1% of the global ASIC market, far behind major contestants such as Bitman and Microbt.
Canaan is adopting a unique strategy when shifting to other AIs, giving priority to mining-centered hardware and strengthening its appearance in markets such as North America. The long -term success of this approach has not yet been determined.
Do you know AI firms face pressure to go green. Bitcoin miners that already use renewable energy like hydro or solar can attract AI customers to meet stability goals through clean colocation deals.
Major risks and ideas for miners entering AI market
As bitcoin miners are rapidly converting to AI, this infection offers opportunity and significant risk. Miners should carefully consider the following:
- Infrastructure cost vs. returns: ASIC-based mining requires adequate initial investment to move to GPU-based AI system. Miners should ensure that potential long -term revenue overtakes these costs.
- Customer stability: AI clients, especially startups, may have frequent funding or lack of long -term reliability. Customers should carefully evaluate to avoid payment of payment or service to the miners.
- Power supply reliability: AI operations demand continuous, high-energy use. Miners should secure stable, long -term power agreements and monitor the local grid capacity to prevent an outage or sudden price increase.
- Cooling and Thermal Management: AI chips, such as Nvidia H100s, produces significant heat. Inadequate cooling system can reduce equipment failures or low efficiency.
- regulatory compliance: Hosting AI workload may include complex rules related to data privacy, intellectual property, international data hosting, energy use, water consumption and carbon emissions. Miners should be prepared to navigate these rules.
- Market competition: As more miners enter the AI collection market, pricing may decline. Early entrances should establish benefits such as strategic space, low energy costs or large -scale operations.
- Resource Stress: Expanding the AI while maintaining mining operations can overstrate financial and management resources.
