By Mari Hansen 
Between reading the statistics and actually sitting with them, there is a point at which the numbers cease to seem abstract. Currently, the worldwide network of Bitcoin miners uses about 173 terawatt-hours of electricity annually. Poland, a nation of 38 million people with factories, hospitals, trains, homes, and everything in between, uses between 150 and 172 TWh a year. The math hits hard but silently.
This is not a forecast for the future. Climate activists did not create this worst-case scenario. In 2025, we are witnessing a financial network based on mathematical puzzles use the same amount of electricity worldwide as a whole mid-sized European country. And the discussion of what that means in terms of morality, practicality, and the environment is just getting started.
| Category | Detail |
|---|---|
| Subject | Bitcoin Mining — Global Energy Consumption |
| Annual Energy Use (2025) | ~173 TWh |
| Comparable Nation | Poland (~150–172 TWh annually) |
| Global Electricity Share | ~0.5% of worldwide demand |
| Continuous Power Draw | ~10 GW |
| Renewable Energy Share | 52.4% (2025) |
| Carbon Per Transaction | ~712 kg CO₂ |
| Energy Per Transaction | ~1,335 kWh (~45 days of a U.S. home) |
| Top Mining Regions | Texas (USA), Sichuan (China), Kazakhstan |
| Regulatory Status | Energy tax proposals under review in U.S. and EU |
| Key Industry Event | Bitcoin Halving — April 20, 2024 (Block 840,000) |
| Reference Website | Cambridge Bitcoin Electricity Consumption Index (CBECI) |
The perception that most people have of these operations is out of date, so it’s important to understand what’s really going on inside. Large-scale Bitcoin mining differs greatly from a hobbyist’s basement setup. Massive server halls are located close to hydroelectric plants in the province of Sichuan, and they are constantly in operation during both the wet and dry seasons.
Similar to how steel mills used to appear in demand forecasts, massive facilities in West Texas have become fixtures on the local utility grid. Because Bitcoin’s presence isn’t subtle, utility engineers in some of these zones have had to completely reconsider load planning.
Currently, a single Bitcoin transaction uses about 1,335 kWh of electricity. For an average American household, that equates to roughly 45 days of power spent verifying a single line in a digital ledger. It is challenging to politely contextualize the number. Following its transition to proof-of-stake, Ethereum reduced its energy consumption per transaction to approximately 35 watt-hours, a decrease of over 99.9%.
Such a shift has not yet occurred in Bitcoin, and there are compelling ideological arguments within its community that suggest it probably won’t. It’s a valid argument. However, it makes it more difficult to rule out the energy question.
It’s true and likely underreported that the industry has made significant strides toward renewable energy. Approximately 52.4% of Bitcoin mining worldwide will use sustainable energy by 2025, up from roughly 37.6% just three years ago. Although estimates vary by region and methodology, coal has decreased to about 8.9% and natural gas makes up an additional 38.2% of the mix.
When grids have excess renewable power and nowhere else to send it, some mining operations have truly reinvented themselves as demand-response tools, increasing consumption only. Helsinki residents have been receiving waste heat from a company called MARA’s mining rigs.
Heat is being directed into greenhouses by a solar-powered mining platform in the Netherlands. These aren’t publicity gimmicks. They are important and genuine engineering solutions.
However, they don’t address the whole issue. About 10 gigawatts are still continuously drawn by the network. Approximately 712 kilograms of CO2 are still released during each transaction. Additionally, its energy appetite continues to grow in tandem with its price, which means that as Bitcoin’s value increases, miners will use more electricity to chase it. Restraint is not a natural outcome of the incentive structure.
The 2024 halving, which occurred on April 20 at block 840,000 and reduced Bitcoin’s block reward from 6.25 to 3.125 BTC, forced an industry-wide reckoning that is still ongoing. Almost overnight, miners using outdated, inefficient equipment were squeezed. In the majority of electricity markets, equipment that consumed thirty or more joules per terahash found it difficult to maintain profitability.
Those who had access to the cheapest power, the newest hardware, or both were the ones who survived. ASICs from the most recent generation now run at about 13.5 joules per terahash, which is about seven times faster than those from 2017.
That increase in efficiency is truly remarkable. Paradoxically, it also makes it more difficult to reduce the network’s overall consumption because cheaper mining attracts more miners, which increases difficulty and necessitates more machines overall.
European and American regulators have taken notice. On both sides of the Atlantic, energy tax proposals for cryptocurrency mining are being drafted and discussed. Three years ago, public mining companies were exempt from disclosing their energy sources and carbon footprints, but now they are under growing pressure to do so.
It’s still unclear if any of these proposals will pass into law with significant enforcement or if, by the time lobbying is finished, they will have been weakened to the point of being ineffective. However, the current direction of travel feels different. The days of just ignoring the question seem to be over.
The fact that Bitcoin’s supporters aren’t wholly incorrect is what complicates this—actually, not rhetorically. In certain situations, the grid stabilization argument is quite persuasive. There is a measurable renewable absorption function.
Mining rigs that turn natural gas into electricity on-site before it escapes can partially offset methane flaring, which releases a powerful greenhouse gas into the atmosphere when oil fields burn off excess natural gas. These are real use cases, not made-up ones.
Over the next ten years, the industry might find ways to strengthen its environmental case. It’s also possible that regulators will act more quickly than the industry can adjust due to pressure from climate commitments and increasingly strained grids. By late 2025, AI infrastructure is expected to equal or surpass Bitcoin’s energy consumption, further complicating the situation.
Suddenly, two massive electricity consumers are vying for the same renewable capacity, and the grid was not designed with either of them in mind.
It’s difficult not to feel that the reckoning is already here as you watch this play out. Now, the question is whether it comes as something more abrupt or as a negotiated transition where industry and regulators find acceptable terms. Poland uses the same amount of electricity as Bitcoin mining. There is more to that fact than just a footnote.