Most people arrive at this question already suspicious. For good reason. This post answers it head-on. Are NFTs bad for the environment, and if so, how bad, measured against what the chain actually consumes today rather than what it consumed in 2021. The short version: the objection had real teeth a few years ago, and then the ground moved. After Ethereum's switch to proof of stake, the energy cost of an on-chain transaction fell by more than 99%. That does not make the footprint zero. It does not let any single project off the hook either. What follows is the deep read on one angle of our environmental NFTs pillar, the part that asks whether the energy cost cancels the good. We are going to keep it grounded. No hand-waving the skeptic away.

Where the bad reputation came from (and why it was fair)

The criticism around 2021 was not noise. It was arithmetic.

Back then, Ethereum ran on proof of work, the same consensus model Bitcoin still uses. Securing the network meant miners racing thousands of specialized machines against each other, burning electricity continuously to win the right to add the next block. That competition was the whole point. It made the chain hard to attack. It also made it enormously power-hungry. At its peak, the Ethereum network drew electricity on the order of a mid-sized country, and every action recorded on it, including minting and trading an NFT, carried a slice of that cost.

So when an environmental NFT was held up as proof that art could fund ecology, the obvious rebuttal landed hard: the medium itself ran on a furnace. The NFT carbon footprint of that era was a fair thing to point at. We are not going to minimize it. Anyone who walked away from NFTs in 2021 over energy was reading the situation correctly at the time.

The trap is assuming the situation stayed frozen there. It didn't.

What proof of stake actually changed

On September 15, 2022, Ethereum executed an upgrade called the Merge. It replaced proof of work with proof of stake, and that single change rewrote the energy math underneath every NFT on the chain.

Here is the mechanism in plain terms. Proof of work secured the network through computation, with machines competing by raw electrical effort. Proof of stake secures it through economic commitment instead. Validators lock up capital as a stake and are selected to confirm blocks, and dishonest behavior costs them that stake. No global race, no warehouses of mining rigs running flat out. The security comes from money at risk, not megawatts burned.

The result was not a trim. It was a collapse. By the Ethereum Foundation's own accounting, the Merge cut the network's annualized electricity consumption by more than 99.9%. A network that once ran like a small country now runs closer to a small town. That is the core of the proof of stake NFT shift, and it is why the headline energy argument from the proof-of-work years no longer describes most NFTs minted today.

A pre- and post-Merge comparison makes the scale clear:

Pre-Merge (proof of work) Post-Merge (proof of stake)
How the chain is secured Miners competing by computation Validators staking capital
Energy driver Continuous electrical effort Money at risk, minimal compute
Network electricity use Country-scale More than 99.9% lower
NFT energy use per action Carried a heavy slice A small fraction of the former cost

For readers who want the underlying numbers and the independent measurements behind them, we break the post-Merge picture down further in NFT energy use after the Merge. The honest framing throughout: much lower, not zero, and estimates vary by who is counting.

"Much less" is not "nothing": the honest accounting

Stopping at the 99% number and declaring the matter closed would be its own kind of dishonesty.

Proof of stake made NFT energy use small per transaction. It did not make it disappear. Validators still run hardware. That hardware still draws power. Independent measurement bodies such as the Crypto Carbon Ratings Institute have studied the post-Merge network specifically, to keep the figure traceable rather than rhetorical. Their work confirms both halves of the truth at once. The drop was real and enormous. The remaining footprint is not zero.

There is also more to an NFT than the moment it is minted. Secondary marketplaces run infrastructure. Image and metadata storage carries a cost. Some NFTs live on chains that never moved to proof of stake at all, and those carry a very different footprint from one on a modern proof-of-stake chain. That part gets skipped by the loudest voices on both sides.

Which is why the cleanest answer to NFT environmental impact in 2026 is that "NFTs" is not one thing. The footprint depends on the chain and the project, not on the three-letter acronym. A blanket verdict, in either direction, is the wrong shape for the question. Ask which chain. Ask what the project does with the value it captures. The answer lives in the specifics.

One more honest note about the figures themselves. The Foundation's own number and an independent body's estimate will not line up to the decimal. They draw the boundary of "the network" in slightly different places and measure at different moments. The Foundation cites a cut above 99.98%, while the commonly repeated round number sits near 99.95%. That spread is normal for a system this size, which is why we cite a range and the source rather than one hard digit. The direction is not in dispute. The exact second decimal is, and pretending otherwise would be the false precision this category exists to avoid.

What a structural 8% allocation is, and is not

This brings the question home to Trash Relics, and it is where we have to be most careful, because this is exactly the point where projects tend to overclaim.

Trash Relics commits a permanent 8% allocation toward real-world environmental preservation and restoration. It is written into the architecture as a fixed structural rule, not a marketing promise renewed each quarter and quietly dropped. That is what it is.

Here is what it is not. It is not an energy offset, and it does not "cancel out" the on-chain footprint. We will not describe it that way. It is not a carbon-neutrality claim, and it is not a guaranteed environmental outcome with a number attached. What it is: a funding commitment. A rule about where a fixed share of value goes. It sits entirely separate from the question of what the chain consumes.

The distinction matters because the easy version of this paragraph would read better and mean less. A project can always say its art "heals the planet" and let the reader fill in the rest. We would rather state the narrow, defensible thing: a fixed share is allocated, by rule, toward preservation and restoration work in the physical world. What that work achieves depends on the work, not on the wording here. Claiming a measured outcome we cannot yet measure would put us right back among the overpromises this whole article is built to reject.

Those two things have to stay uncoupled to stay honest. The low on-chain footprint is a property of proof-of-stake infrastructure. The 8% is a property of how this archive is built. One does not pay off the other. Treating a funding rule as if it neutralized emissions is the precise move this article exists to refuse. The mechanism, and exactly how the allocation is structured, is documented on the Environmental Impact page, which owns that detail in full.

Short answers to the common follow-ups

Do NFTs still use as much energy as Bitcoin?
No. Bitcoin still runs on proof of work, the high-energy model. NFTs on a proof-of-stake chain like post-Merge Ethereum run on a fundamentally different consensus mechanism with more than 99.9% lower network energy use. The comparison that held in 2021 no longer holds.

Does the 8% allocation make Trash Relics carbon neutral?
No, and we would not claim it does. The 8% is a structural funding rule directed at real-world preservation, separate from the chain's energy footprint. It is a commitment to where value goes, not an offset and not a neutrality claim.

Why does the chain a project uses matter so much?
Because the consensus mechanism, not the file, determines the energy cost. Two NFTs can look identical and carry very different footprints depending on whether their chain runs proof of work or proof of stake. When you assess any project on environmental grounds, the chain is the first thing to check.

Where this leaves the question

So, are NFTs bad for the environment? The grounded answer is that the footprint is real, far smaller than the headline that made the question famous, and dependent on the chain and the project rather than on NFTs as a category. The 2021 criticism was fair for 2021. It is a poor description of a proof-of-stake NFT in 2026.

Trash Relics sits inside that reality without pretending to escape it. The on-chain footprint is small because of the infrastructure. The 8% allocation is a fixed rule, not a rescue, and not a cancellation of anything. Keeping those two claims apart is the whole discipline here.

The full mechanism is documented on the Environmental Impact page, and the wider picture of art that funds ecology sits in the environmental NFTs pillar. The object remains evidence, not an offset.

Sources

  • Ethereum Foundation, "Ethereum energy consumption" — ethereum.org/en/energy-consumption (reduction of more than 99.988%, per CCRI)
  • Ethereum Foundation, "The Merge" roadmap — ethereum.org/en/roadmap/merge (date: September 15, 2022; ~99.95% energy reduction)
  • Crypto Carbon Ratings Institute (CCRI), "The Merge — Implications on the Electricity Consumption and Carbon Footprint of the Ethereum Network" — carbon-ratings.com