Bitcoin mining’s annualized electricity demand rose to about 190 terawatt-hours in December 2025, up 38% from 138 the in June 2024, according to preliminary research reported by theEnergyMag.
Summary
- Bitcoin mining electricity use rose 38% to 190 TWh between June 2024 and December 2025.
- Hydropower became mining’s largest energy source as low-carbon power reached 59.4% of the reported mix.
- Only 10% of surveyed miners had already allocated power to AI or accelerated computing services.
Alexander Neumueller of the Cambridge Centre for Alternative Finance presented the figures at the Energy Investors Forum in Dallas. Cambridge expects to publish the second edition of its Digital Mining Industry Report later in 2026.
The research also found that hydropower had overtaken natural gas as Bitcoin mining’s largest single energy source. Low-carbon power supplied 59.4% of the reported mining mix, up from 52.4% in the previous study. However, total estimated greenhouse-gas emissions still increased by 20%, from about 40 million to 48 million tonnes of carbon-dioxide equivalent.
Hydropower takes the largest share of mining power
The 2025 Cambridge Digital Mining Industry Report found that natural gas supplied 38.2% of surveyed miners’ electricity, making it the largest single source at the time. Renewables provided 42.6% in total, while nuclear power added 9.8%. Coal’s share had fallen to 8.9%, down from 36.6% in the earlier 2022 estimate.
The preliminary update changes that order. Hydropower now ranks ahead of natural gas, although Cambridge has not released the full breakdown for each source. Neumueller linked part of the change to stronger survey coverage in hydro-rich markets such as Ethiopia.Ethiopia expanded Bitcoin mining around low-cost electricity from the Grand Ethiopian Renaissance Dam.
Electricity demand rises faster than emissions
The network’s annualized power use increased by about 52 TWh between the two reference points. Annualized demand measures the electricity Bitcoin mining would use over a year if the December 2025 rate continued. It does not mean miners consumed exactly 190 TWh during the 2025 calendar year.
Emissions rose more slowly than electricity demand because miners reported using a lower-carbon power mix. Even so, Cambridge’s estimate still increased from roughly 40 million to 48 million tonnes of CO₂ equivalent. The cleaner mix slowed the rate of emissions growth, but it did not offset higher overall electricity consumption.
More mining machines joined the network during the measured period, raising total computing power. Newer hardware can perform more calculations for each unit of electricity, but efficiency gains did not fully counter the increase in hashrate. Cambridge’s Bitcoin Electricity Consumption Index tracks how prices, transaction fees, mining equipment and network difficulty can change estimated electricity demand over time.
Preliminary figures carry survey limits
Cambridge based the new estimates mainly on responses from mining companies representing slightly more than half of global Bitcoin hashrate. The wider coverage gives researchers a larger sample than the first report. However, the final publication may revise some figures after Cambridge completes further checks.
The 2025 report also warned that survey participation can distort geographic estimates. U.S. companies supplied a large share of responses, which likely overstated the country’s portion of global mining activity. The latest rise in reported hydropower may partly reflect better coverage of miners in Ethiopia and other markets that rely more heavily on hydroelectric generation.
Cambridge’s earlier study estimated 39.8 million tonnes of emissions using its survey-based method. A separate location-based model produced a much higher estimate of 69.6 million tonnes. The gap shows that results depend on assumptions about mining locations, electricity contracts, grid mixes and the use of stranded or flared energy.
Miners explore AI, but deployments remain limited
The new survey also examined whether Bitcoin miners are shifting power capacity into artificial intelligence and high-performance computing. About 10% of respondents said they had already allocated some power to AI or accelerated computing. More than 40% of the remaining miners said they were actively exploring the option.
Neumueller cautioned that “intent to look into it is not commitment to deploy.” AI data centers need costly networking, cooling and reliability systems that basic Bitcoin mining sites may not have. Miners can quickly reduce Bitcoin loads when electricity prices rise, while AI customers usually require steady power and stronger service guarantees.
Still, almost nine in ten respondents expected AI and HPC diversification to gain ground over the next several years. As crypto.news reported, listed miners have already announced more than $70 billion in AI and HPC contracts as they seek steadier revenue outside Bitcoin production.
The change is already visible in some company results. TeraWulf generated more revenue from HPC hosting than Bitcoin mining during the first quarter of 2026. It reported $21 million from HPC services, compared with less than $13 million from digital asset mining.
The Cambridge findings show two changes taking place together. Bitcoin mining uses more electricity, while hydropower and other low-carbon sources account for a larger share. At the same time, mining companies are assessing whether their power connections and sites can support AI services. Cambridge’s full report will provide a detailed energy breakdown and final methodology later in 2026.
