The Subtle Way AI Data Centers Boost Bitcoin Mining Economics

As a seasoned researcher with a keen interest in the intersection of technology and economics, I find the potential impact of AI data centers on the bitcoin mining industry intriguing. From my perspective, the increased competition for cheap electricity between these two sectors could indeed establish a floor for hashprice, benefiting miners in the long run.

The growth of AI-focused data centers could potentially benefit Bitcoin mining economics, not just those involved with AI, in an easy-to-understand sense.

The rivalry between AI data facilities and Bitcoin miners over affordable electricity might ultimately determine the minimum price of hashrate, an essential indicator that miners use to gauge their income.

Spencer Marr, president of bitcoin mining company Sangha Renewables, told CoinDesk that each prospective mining investment is now being evaluated in this manner: should this location be utilized for AI applications or mining? Whenever they opt for AI or other advanced computing methods, the total hashrate and hash price won’t increase, thereby avoiding a negative impact.

In simple terms, the hash rate represents the total computing power that supports a Proof-of-Work blockchain like Bitcoin. On the other hand, hash price is the amount of bitcoin a miner receives when their machines complete a specific number of computations or hashes within a defined time period.

Currently, the combined computing power of bitcoin mining, or its hashrate, stands at approximately 770 exahashes per second (EH/s), according to Hashrate Index. The cost per unit of this collective mining power, known as hashprice, is around $61.12 per petahash per day. Over time, the hashprice has been gradually decreasing due to increasing competition among miners. Compared to 2017, when it wasn’t unusual for the hashprice to surpass $1,000 by this measure.

Setting a minimum price for hashing (or mining) operations would be beneficial for miners because it would ensure their computing power remains above a specific value floor, regardless of the prevailing conditions.

As a crypto investor, I’ve noticed that in the race to secure cheap electricity, miners are increasingly finding themselves at a disadvantage compared to more eager buyers who are other types of computational systems. This is a unique application of game theory, as a miner, I prefer it when others decide to use these affordable energy resources for purposes other than Bitcoin mining due to the inherently competitive nature of this field. In simpler terms, I’m saying that if more people use electricity for tasks other than Bitcoin mining, it makes it easier for me to mine Bitcoin because there’s less competition for the same limited resource.

However, the pressure could cause Bitcoin miners to relocate their operations to different countries worldwide, where AI data centers are less prevalent, as suggested by Jaran Mellerud, co-founder of Hashlabs Mining, a company specializing in Bitcoin mining hardware and hosting services, during an interview with CoinDesk.

“Mellerud believes that the struggle for power among AI centers won’t greatly affect Bitcoin mining costs because the Bitcoin mining system naturally adjusts itself. If the mining rate decreases in one nation, it will boost the earnings of miners in other countries, enabling them to expand further. Mellerud also predicts that by 2030, less than 20% of the total Bitcoin mining rate will be controlled by the U.S., while there will be significant growth in mining activity in Africa and Southeast Asia.

Marr acknowledged these arguments were sound, but he added a caveat: “Ultimately, there’s a limited supply of extremely affordable electrons.” Operating AI data centers is more complex than managing bitcoin mines; they demand continuous operation, for instance, and their construction and maintenance costs are significantly higher.

Marr suggested that while the contest for electrons might eventually slow, it won’t necessarily halt the increase in hashrate.