With the rapid development of artificial intelligence technology, the demand for computing infrastructure continues to climb, while Bitcoin mining faces the dual pressures of rising computing difficulty and increasing electricity costs. This has led to the view that converting idle mining farms into AI data centers may be a way out of the industry's predicament. However, Interhash CEO Alexander Lozben points out that this idea ignores the structural obstacles in actual operation, especially in the Russian market, where such a conversion is far from a simple replacement of equipment.
The core contradiction at present is not the competition between mining and AI, but the competition between the two for limited power resources. Whether it is a large Bitcoin mine or an AI cluster for training large models, the power demand of a single facility is generally between 20 and 100 megawatts. However, the regions in Russia with surplus power capacity are extremely limited, and the approval cycle for new grid access often takes months or even longer. Therefore, the key to determining the feasibility of a project is no longer the price of electricity, but whether a stable and sufficient supply of electricity can be obtained.
Even if the energy problem is solved, the differences in hardware infrastructure constitute a difficult-to-cross divide. Bitcoin mining relies on dedicated ASIC chips, focusing on high-density deployment and energy efficiency; while AI data centers require high-performance GPUs and accelerators, with much higher requirements for heat dissipation, power supply stability, network bandwidth, and operational precision than mining machines. AI data center standards include redundant power systems, precision temperature control, and low-latency network architectures, almost all of which are lacking in traditional mining farms.
The gap in capital investment is even more significant. The cost of building a mining facility per megawatt is about $300,000 to $600,000, while an AI data center of the same size requires $3 million to $6 million, more than ten times the difference. This means that the so-called "transformation" is almost a reconstruction, with costs and cycles far exceeding expectations.
In terms of business model, the two are also very different. Bitcoin mining relies on a protocol mechanism, with revenue determined by block rewards and transaction fees. The system automatically adjusts the difficulty and does not rely on external customers, making it more resistant to cycles. The revenue of AI data centers, on the other hand, relies entirely on long-term contracts with tech giants such as Microsoft, Google, Amazon, and Meta. In Russia, the number of such corporate clients is small, and market demand has not yet formed a scale, making it difficult to establish a commercial closed loop.
Although some companies around the world claim to have completed the transformation from mining farms to AI centers, a deeper analysis reveals that the vast majority of cases do not involve the reuse of equipment, but rather the relocation and construction based on the original power resources. True "transformation" is still in the minority and highly dependent on policy support and capital inclination.