In March 2021, an artist who called himself Beeple sold a JPEG at Christie's for $69 million. The buyer didn't get a painting, a print, or even an exclusive copy. He got a token on a blockchain — a receipt — pointing to a file anyone with an internet connection could right-click and save.
That sale, more than any other, kicked off the NFT era. For about eighteen months, digital ownership was the future. Bored apes and CryptoPunks filled Twitter avatars. Celebrities minted collections. Galleries opened virtual showrooms. The promise was that scarcity, finally enforced by cryptography, would do for digital art what auctions had done for oil on canvas.
Five years later, the numbers tell a different story.
The collapse
By the first quarter of 2025, art NFT trading had dropped to roughly $23.8 million — down from $2.9 billion at the 2021 peak. That's a 93% collapse, according to data tracked by DappRadar. Active art-NFT traders fell from more than 529,000 in 2022 to fewer than 20,000 in early 2025, a 96% decline. The twenty most-traded collections lost about 95% of their volume and sales.
The platforms followed. KnownOrigin and MakersPlace — pillars of the early NFT art world — shut down. SuperRare and Art Blocks, the prestige players, lost more than 90% of their volume. Foundation, once a tastemaker for one-of-one drops, went nearly silent.
This wasn't only an art problem. Total NFT trading across categories fell to its lowest levels since 2020. By 2024, monthly NFT volume had dropped roughly 86% from its peak, and total NFT market cap had been cut in half. Even the maximalist crypto press had stopped pretending: 2024 was widely described as the worst year for NFTs since the technology became famous.
What collapsed was not the technology. NFTs are still being minted; the underlying primitive — a unique token recorded on a public ledger — works exactly as designed. What collapsed was the cultural premise. The premise that artificial scarcity, imposed on infinitely copyable files, was a thing the internet actually wanted.
Meanwhile, in the commons
While Christie's was selling JPEGs for the price of a townhouse, a quieter project was passing milestones almost no one wrote about.
Creative Commons launched its first set of licenses in December 2002. Twenty years in, the catalog of works released under those licenses had passed two and a half billion. That's billion, with a b. Photographs, scientific papers, sheet music, government datasets, video, audio, encyclopedia articles, classroom curricula, museum collections — all openly shared, attributed, remixable, and free.
Every Wikipedia article you've read since 2009 is CC-licensed. So is most of Flickr's open archive. So are the 2.8 million images and datasets the Smithsonian released into the public domain. NASA's most iconic photographs. A growing share of academic papers across PLOS, eLife, and the open-access journal landscape. Most of the world's open educational resources. The Free Music Archive. Wikimedia Commons. Openverse. Europeana. The Met's open collection.
CC works are how museums share their holdings with the public. They are how teachers in low-income countries get textbooks. They are how the Wikipedia article you'll read tomorrow is even legally allowed to exist. They are how a developer in Lagos can legally use a photograph taken in Helsinki for a project funded in São Paulo.
In the same window during which NFT volume collapsed by 93%, the commons just kept compounding.
Why one survived and the other didn't
NFTs and Creative Commons started from opposite premises about what the internet is for.
NFTs took the model of physical scarcity — one painting, one owner — and tried to graft it onto digital files using a blockchain. The goal was to make digital art behave like physical art: unique, ownable, expensive, investable. The technology delivered the artificial scarcity, but the culture never agreed to honor it. Right-clicking remained legal, free, and morally uncontroversial. The market that emerged was, accordingly, mostly speculation. People bought tokens not because they loved the image but because they expected someone else to buy the token from them later for more. When that expectation stopped, so did the market.
Creative Commons started from the opposite premise: that copying is what computers and networks naturally do, and the law should be reshaped around that fact rather than fight against it. CC licenses don't try to make digital files behave like physical objects. They take digital abundance for granted and ask a different question: under what terms is the creator happy for this work to spread?
That question scales. The more a CC-licensed work is copied and shared, the more it succeeds. Wikipedia is more valuable because everyone has it. The Smithsonian's open archive is more valuable because every classroom can use it. A CC-BY photograph on Flickr does its job better the more it gets reused. There's no floor price to defend, no liquidity to maintain, no greater fool required.
NFTs needed the next buyer. The commons just needs the next user.
What infrastructure looks like
A useful way to think about this is the difference between speculation and infrastructure.
Speculation is loud. It generates headlines, celebrities, conferences, podcast episodes, fortunes, and ruin. It runs on attention and capital flows. When attention moves on, the value evaporates.
Infrastructure is quiet. It is the thing other things are built on. Roads, plumbing, DNS, HTTP, SMTP — the systems with the highest cumulative impact tend to be the ones nobody talks about. They become invisible because they work. Creative Commons is internet infrastructure for sharing. It became invisible the same way the URL bar became invisible: by being everywhere, by working, and by costing nothing to use.
You can measure this asymmetry directly. The peak NFT art market was about $2.9 billion in annual trading — a real number, but a number measured in transactions, not in cultural use. Two and a half billion CC-licensed works, by contrast, are not "trading volume." They are the ambient material of contemporary digital culture. Every Wikipedia search, every open-access paper read, every museum image embedded in a school project draws on them. Their value isn't in the price of the next sale. It's in the absence of friction.
What survives
History is reasonably consistent about which kinds of digital projects last. The ones designed for ownership and exclusivity tend to peak with their hype cycle and decay with it. The ones designed to lower the cost of sharing tend to compound, because every new participant makes the system slightly more useful for everyone else.
The NFT era will be remembered. It changed how some artists got paid for a brief window. It pushed conversations about provenance and digital rights into the mainstream. Some of the underlying technology will find real uses in identity, ticketing, and gaming. None of that is nothing.
But it didn't outlast Creative Commons, and it isn't going to. The commons is older, larger by orders of magnitude, growing faster, and built on a premise the internet actually rewards. A receipt for a JPEG was always a strange thing to want. A way to legally share that JPEG with the entire world, attributed and unrestricted — that was always the more interesting idea.
Catalogs of openly licensed media exist for a reason. They are how people find work that is already free to use, already part of the shared cultural record, already designed for the medium it lives in. While the speculative web spent five years promising digital ownership and delivering volatility, the commons spent the same five years doing what it has been doing since 2002: getting bigger, and making more things possible.
That's what won.