Digital Art and the Open Commons — What Generative AI, NFTs, and the Creative Commons Movement Are Doing to Each Other

Digital art has always had a complicated relationship with ownership. A painting exists once, in a specific physical location, and its ownership is unambiguous. A digital image exists in as many copies as have ever been made of it, identically, simultaneously, in locations that may not even be determinable. The entire history of digital culture is, in one reading, an extended negotiation between the logic of the digital medium — perfect, free, infinite replication — and the economic and cultural systems built on scarcity and control.

Three developments in 2026 are making this negotiation more intense and more consequential than at any previous moment. Generative AI has made the production of sophisticated visual art accessible to anyone with a prompt. Non-fungible tokens — NFTs — have created a mechanism for asserting scarcity and ownership within the otherwise infinitely replicable digital medium. And the Creative Commons movement, which has spent twenty-five years building the infrastructure for open sharing of digital culture, is trying to understand what its role is in a landscape that looks fundamentally different from the one it was designed for.

What Generative AI Did to Digital Art

The emergence of practical image generation from text prompts — beginning with DALL-E and Midjourney in 2022 and accelerating rapidly through 2023-2025 — did not merely add a new tool to the digital artist's toolkit. It introduced a new kind of object into the cultural economy. Images that would have required professional artistic skill, expensive software, and significant time to produce can now be generated in seconds by anyone who can describe what they want.

This has at least three significant effects on the creative commons.

The first is on authorship and the question of who owns what gets generated. In most jurisdictions, copyright requires human authorship — a work produced entirely by a machine without human creative input is not eligible for copyright protection. The US Copyright Office has taken a clear position: works generated entirely by AI cannot be registered for copyright. This means that a significant category of image now exists that is, in the most direct sense, in the public domain from the moment of creation. It belongs to no one. Anyone can use it for any purpose.

For the commons, this is a genuinely ambiguous development. The public domain grows by default when AI generates uncopyrightable images — a vast pool of images without copyright holders is, in principle, available for anyone to use. But the utility of that pool depends on what the images are of, and on whether the absence of copyright is practically meaningful when the images are generated by proprietary systems that control access to their tools. The image itself may be public domain; the ability to generate more images like it is controlled by a private company.

The second effect is on the market for human-made digital art. Generative AI has compressed the price point for competent visual execution dramatically. Illustrations that would have cost hundreds or thousands of dollars to commission from human artists can now be produced for pennies through AI services. This is not a universally negative development — it has made access to high-quality visual content more democratic — but it has significantly disrupted the economic foundations of digital illustration, stock photography, and commercial visual art.

The third effect, directly relevant to the commons, is the training data problem described in other coverage on this platform. The generative AI systems that produce these images were trained on billions of images scraped from the internet — including, in many cases, images released under CC licences and images that were part of the cultural commons by any reasonable definition. The creators of those training images had no say in their inclusion. The systems that were built on their creative labour are now competing with them commercially.

What NFTs Actually Established — and What They Did Not

The NFT phenomenon of 2021-2023 was widely caricatured and just as widely misunderstood. At its height, NFTs were described as either a revolutionary new model for artist ownership and monetisation or an obvious speculative bubble of no lasting cultural significance. Both characterisations missed what was actually interesting about them.

What NFTs actually established, at least as a technical proof of concept, was the possibility of tracking provenance and establishing a chain of ownership for a digital file on a public ledger. The NFT itself is not the artwork — it is a record, stored on a blockchain, that asserts a specific file as the canonical instance of a specific work, and that tracks who has held that record. The artwork is whatever the file contains. The NFT is a certificate.

The cultural significance of this certificate was never primarily about scarcity in the sense that opponents of NFTs claimed — the file can still be copied, the image can still be downloaded and reposted. The significance was about social consensus around provenance. In a medium where everything is infinitely copyable, NFTs created a mechanism for a community to agree that this particular instance, held by this particular person, is the one that counts for the purpose of ownership and value.

For the commons, NFTs raised a specific and underexplored question. What is the relationship between CC licensing and NFT-based ownership claims? If an artist releases a work under CC BY and later mints an NFT asserting exclusive ownership of that work, what has happened to the licence? The answer under current understanding is that the CC licence remains in effect — it was irrevocably granted to all users who received the work under its terms. The NFT holder owns the NFT and whatever social value the community attaches to that ownership. They do not own exclusive rights to the underlying work if it was previously CC-licensed.

This tension has produced real disputes. Artists who discovered their work minted as NFTs without permission — a practice sometimes called "right-clicker mentality" in NFT culture — found that the CC licences they had applied to their work created ambiguity about whether the minting was even a licence violation. An NFT mint is arguably a transformation of the work, which might require a CC BY-SA licence to remain open, or might require an ND licence to prevent — but the legal analysis is not settled.

Where the Three Forces Intersect

The intersection of generative AI, NFTs, and the Creative Commons is where the most interesting and unsettled questions about digital art live in 2026.

Generative AI has created vast quantities of uncopyrightable images. Some of those images are being minted as NFTs and sold, which asserts an ownership claim over objects that are legally in the public domain. The legal status of this practice is uncertain — you can sell an NFT of a public domain work (museums have done this with digitisations of historical art), but the NFT holder acquires no copyright in the underlying work. Anyone can use the image freely regardless of who holds the NFT.

CC-licensed works have been used as training data for generative AI systems without systematic creator consent. Some of those systems now produce outputs that closely resemble specific CC-licensed artists' styles. The CC licences do not specifically address this — the question of whether style can be protected is a separate copyright question from whether the training process itself required a licence.

Artists who have built their practice around CC licensing and open culture principles find themselves navigating all of this without clear guidance from the institutions — Creative Commons, copyright law, platform terms of service — that are supposed to provide it.

What Open Digital Art Culture Looks Like When It Works

The picture above is complicated and in some ways troubling, but the commons has resources and a tradition that are worth naming clearly.

There is a long and productive history of digital art that has been made freely available, built upon, and extended in ways that created genuine cultural value. The demo scene — a community of programmers and visual artists who have created elaborate audio-visual demonstrations since the 1980s and who have always shared their work freely — predates Creative Commons by decades and has produced some of the most technically remarkable digital art in existence. Open source visual art tools — Blender, GIMP, Inkscape, Krita — have been built by communities that operate on the same principles as the software commons, and have produced professional-grade creative tools available to anyone at no cost. The Flickr Commons, which makes museum and library photography collections available under open licences, has given digital artists access to a vast archive of historical images that enriches creative work across the internet.

These traditions did not require NFTs to establish value, did not depend on AI to generate content, and did not face copyright enclosure by commercial platforms. They worked because the communities involved understood that shared culture is more valuable than enclosed culture, and because they built the infrastructure to make sharing practically possible.

The challenge for digital art and the commons in 2026 is to preserve and extend that understanding in an environment where the pressures toward enclosure are stronger, more technically sophisticated, and more commercially powerful than they have ever been. Generative AI can either be a force that degrades the commons — by training on open culture without reciprocity and competing with the creators who made that openness possible — or a force that extends it, if the tools are made accessible, if the training data practices are made ethical, and if the governance of AI development includes the voices of the open culture communities whose work made the systems possible.

The same is true of NFTs and blockchain-based ownership mechanisms. They can be tools for enclosing digital culture — for asserting private ownership over works that were previously shared — or they can be tools for recognising and compensating creators who have contributed to the open commons, if the social and technical infrastructure is built with that goal in mind.

What determines the outcome is not the technology. It is the values and priorities of the communities that build with it. The Creative Commons movement has spent twenty-five years demonstrating that it is possible to build digital culture infrastructure that serves the commons rather than the individual. That work is more relevant in 2026 than it has ever been, and more urgently needed.