Property rights have been fundamental to Western philosophy and politics since the Enlightenment. When the Web emerged, it dramatically reconfigured this landscape. As cyberspace continues to evolve, the internet of 2024 looks very different from just one or two decades ago - and this isn't solely due to technological progress. These historic shifts can largely be characterized through changing property relations, from Web2 platform capitalism to Web3 NFTs to today's AI revolution. The history of the digital realm is, at its core, a history of intellectual property.
What Even Is Property, Really?
Property is complex and often nebulous. If I pick up a pebble from a public beach and take it home, can I claim ownership? If I don't own it, who does? Does anyone?
Our modern concept of property stems largely from John Locke's Labour Theory of Property. As he wrote in 1689:
"Though the earth...be common to all men, yet every man has a property in his own person: this nobody has any right to but himself. The labour of his body, and the work of his hands, we may say, are properly his. Whatsoever then he removes out of the state that nature hath provided, and left it in, he hath mixed his labour with, and joined to it something that is his own, and thereby makes it his property."
In the case of our beach pebble, it's held in common by everyone until labour is applied — if I draw a face on it or carve it into a new shape, I can claim some ownership through that work. What we then do with that property, and how it interacts with others' property, forms what legal scholars like Kevin Gray think of as a bundle of rights:
‘When I sell you a quantum of airspace the whole point is that – apart from molecules of thin air – there is absolutely nothing there. The key is, of course, that I have transferred to you not a thing but a “bundle of rights”, and it is the “bundle of rights” that comprises the property.’
Anthropologist David Graeber follows this view that property is ‘thingless’, viewing it in terms of social relations;
“One might argue that property is a social relationship; when one buys a car one is not purchasing the right to use it so much as the right to prevent others from using it-or, to be even more precise, one is purchasing their recognition that one has the right to do so. But since it is so diffuse, a social relation- a contract, in effect, between the owner and everyone else in the entire world, it is easier to think of it as a ‘thing’”
The formal construction of social relations is the realm of economics; property can be bought, sold, traded, rented and transferred. As a result, markets are central to our conceptions of property. Objects have market value, determined in neoclassical economic formations by the demand relative to the supply. As such, scarcity is valuable.
This brings us to intellectual property — the attempt to apply these same market-based ownership concepts to creations of the mind.
The Intellectual Property Mess
Copyright covers everything from books and music to computer programs and technical drawings. But there are serious flaws in trying to apply tangible property rights to intangible ideas. New expressions and ideas are rarely (if ever) conjured up from whole cloth; instead, they are the result of a vast array of informational inputs. We all stand on the shoulders of giants. As Cory Doctorow critiques:
"The labour theory of property always begins with an act of erasure: 'All the people who created, used, and improved this thing before me were doing something banal and unimportant – but my contribution is the step that moved this thing from a useless, unregarded commons to a special, proprietary, finished good.'"
Markets are inherent to our capitalist formations of property rights; this is no different for intellectual property. The legal and political frameworks that are employed in the construction of intellectual property regimes necessarily assume that creation and monetisation go hand in hand; works are created with the intent to sell or rent. As a result, the ‘bundle of rights’ that accompany intellectual property are intimately connected to economics.
Imagine that you have purchased a vinyl recording of a particular piece of music. The owner of the rights to that piece (the label or recording artist) has sold the recording to you, but not the rights to the song itself. You can listen to it at home, but you cannot play it in public; You can sell the record, and you can even rent it out, but you cannot make a copy of it; even if you are the only person who intends to listen to the copy.
The object is yours to do with as you please, but the intangible content of the object, the song itself, is not.
The centrality of markets to intellectual property has important political and legal ramifications; it is not the primary goal of intellectual property regimes to proliferate ideas or creative works or to expand access to knowledge. Since value is found in scarcity, not ubiquity, intellectual property regimes are regimes of control over supply, limiting the transmission and sharing of knowledge and content, thereby providing market-based incentives to create new content, or the very limited ability to use existing content in new ways.
“Property rights, of course, confer control as perhaps their most basic feature: the ability of the rightsholder to determine the use of the property. Yet this characteristic places intellectual property in fundamental tension with the underlying goal of creating a source of open, common information (and thus a font of creativity and invention). Too much control… turns the public good into merely a private benefit. The public domain withers, and progress slows.”
As legal scholar Keith Aoki notes, this runs "directly counter to understandings of early English and American copyright law, which prohibited unauthorized copying of texts precisely to promote wider circulation of the ideas, knowledge, and information contained within literary work."
Clearly, technology is deeply intertwined with intellectual property as a concept; Whether the printing press, vinyl records or the internet, almost all communication and information technology can be described as tools that transform the supply, increase the access, or improve the quality of copies of intellectual property. The forces of communication technologies therefore sit directly at odds with the market forces that seek to exert control over the supply of property.
Unsurprisingly, communication technologies have consistently run up against capitalist formations that protect rights holders by proliferating works more widely and more cheaply than the law has accounted for; as new technologies appear, so too have new laws and systems designed to protect rights holders and claw back control over work and works.
Since the physical world necessarily imposes limitations on intellectual property proliferation, Market forces have historically been aided by physics; On the one hand, printing a book or a vinyl record or any other form of physical media, even at unimaginable scale, is limited by cost, time, and resources. Eventually, the paper and plastic and money required to print these objects will run out. Utimately, the illegal reproduction of copyrighted works is stymied by entropy; a copy of a copy of a copy of a copy will necessarily be of worse quality than the original.
Cyberspace, however, has no such limitations.
When Property Goes Digital
I like to think of the history of the internet as a battle of ideas and ideals. The early internet embodied a particular set of values held by people like Richard Stallman:
"I believe that all generally useful information should be free. By 'free' I am not referring to price, but rather to the freedom to copy the information and to adapt it to one's own uses... When information is generally useful, redistributing it makes humanity wealthier, no matter who is distributing and no matter who is receiving."
Christian Sandvig’s seminal paper The Internet as the Anti-Television points out that the internet didn’t have to be this way:
It is built on a set of ideological and utopian visions about the nature of information and access to knowledge. Personalization was assumed; users would all want different information. The early Internet was to be a network of equals, with the ideal user thought to be producing new knowledge, not just passively receiving it… In this future network users would also be producers, content would be plentiful, and attention to it would be widely distributed”
Every internet user's role as a consumer, sharer, and producer of digital content is affected by and in turn affects the technical and organisation of the services they use. In this way, network architecture serves as a form of internet governance; by changing the design of the networks underpinning internet-based services, as well as the global internet itself, the politics of the internet change too, transforming the balance of rights between users and providers, influencing the capacity of online communities to engage in open and direct interaction, and impacting fair competition between actors in the internet market.
The utopian vision set out in the early days of the internet created huge problems for rightsholders; digital files can be infinitely copied without degradation or meaningful cost; the internet has dramatically increased the opportunities and methods for publication of digital content, while networks have made distribution simple and commonplace. This was intentional, and by design.
But as the internet matured, this utopian vision gave way to new systems of control. Chief among these was Digital Rights Management, a set of security technologies that emphasised encryption and authentication to control content consumption and usage. Digital watermarking was used to embed a unique identifier into the content that can be used to track usage of the content; access control and usage control mechanisms attempted to limit the number of times a user can access the content. These systems were largely ineffective; they are onerous to use, alienating the consumer, and created a cat and mouse game with teams of phreaks and hackers who saw new forms of DRM as a challenge to be conquered in the name of free information.
In the end, piracy was defeated not by making it harder to steal through the application of DRM, but by making it easy to consume media through the rise of subscription services. Netflix, Spotify, and a raft of similar platforms shifted the focus away from consumers owning intellectual property, as one would with a DVD, to viewing content on demand for a fixed monthly fee. This is what we have come to call Web2.0: The point-to-point system of equals that defined the early internet was willfully bent toward the older commercial vision of mass communication. Dramatically increasing the size and use of data centres, and the emergence of protocols such as RealAudio, changed the transmission of data away from file sharing and towards streaming. This streaming enabled the collection of rents; SaaS platforms and subscriptions services are rent seeking; extracting wealth from ownership of intellectual property as a capital asset. As Marx observed, long before the emergence of Big Tech, rentier economies concentrate wealth in the hands of the firms and individuals that own capital assets, and the same is true within the digital space. These models have made these platforms extraordinarily wealthy and concentrated power in the form of tech monopolies.
Web3: A Failed Revolution
Blockchain technology and NFTs emerged as a response to Big Tech's dominance. The promise was compelling - using distributed ledgers to enable "true" digital ownership, making digital objects both perfectly copyable AND provably unique through tokenization.
NFTs correctly recognized that scarcity drives value in intellectual property markets, and that ownership could counter rent-seeking. They aimed to let digital creators sell scarce works outside Big Tech's control while giving buyers permanent proof of ownership.
But NFTs fundamentally misdiagnosed the problem. The "scarcity" only applies to the token itself, not the underlying digital asset. There's no cryptographic link between token and asset - the file can be changed server-side at any time. There's no proof of authenticity or confirmation that the minter actually created or owns the original work.
Most damningly, NFTs are still just another market in intellectual property, governed by the same logic that prioritizes trading over creation. Ultimately, NFTs failed to challenge the system.
AI: The Final Boss Battle for IP
Now, Generative AI has arrived to throw all our frameworks into chaos. These deep learning systems ingest vast amounts of existing content to create new, plausibly original works. This raises thorny questions about ownership and rights.
Consider the following scenarios:
Scenario 1
Alice creates a body of unique, copyrighted work that is ingested into an AI generator without their consent. Users can therefore utilise the AI to create a novel work with a prompt such as ‘create a picture in the style of Alice’.
Scenario 2
Bob creates a body of unique work in the style of Alice’s copyrighted work, and gives permission for it to be ingested into the AI generator. Users can utilise the AI to create a novel work with a prompt such as ‘create a picture in the style of Alice’.
Copyright law affords artists and creators no stylistic protections in either case; The US Copyright office is explicit in its assertion that copyright does not protect “Ideas, procedures, methods, systems, processes, concepts, principles, or discoveries” on the one hand, and “works that are not fixed in a tangible form” on the other. Style falls safely into these two categories. This has been upheld by the court on numerous occasions; in Dave Grossman Designs v. Bortin, the court said that:
"The law of copyright is clear that only specific expressions of an idea may be copyrighted, that other parties may copy that idea, but that other parties may not copy that specific expression of the idea or portions thereof. Any artist… may paint a picture of any subject in the Cubist motif, including a portrait of three women, and not violate Picasso's copyright so long as the second artist does not substantially copy Picasso's specific expression of his idea." (Dave Grossman Designs, Inc. v. Bortin, 1972)
Based on these rulings alone, neither scenario has infringed on Alice’s rights as an artist in any way; and yet, it feels grossly unfair to the artist and her work.
There is a second question here; who owns the works produced by AI? According to OpenAI’s terms of service, the input (the prompt) is owned by the user, assuming that it is copyrightable in the first place. Further, the user owns the output from that prompt; OpenAI ‘assigns to you all its right, title and interest in and to Output. This means you can use Content for any purpose, including commercial purposes such as sale or publication."
The sheer size and scale of the datasets at play within Generative AI is not only astonishing, it means that it touches almost everyone who has ever published anything online; GPT2, an earlier iteration of Open AI’s Large Language Model, was trained on the entire database of Reddit posts and comments. Whilst Redditors own the content they create on the platform, the licence agreement that they enter into when using the platform gives Reddit broad powers to use that information however they see fit, up to and including sharing it with OpenAI. It is not just Reddit; one AI researcher estimates that we are one standard deviation and two iterations away from running out of data for Generative AI. In a sense, Generative AI is the great leveller; we are all contributors to the novel ideas that these platforms generate.
Time for a Reset?
Our current intellectual property frameworks are clearly unfit for the AI age, as I've explored in previous posts; rather than protect creators' rights, the government is likely to shield platforms from liability while placing responsibility on end users - just as Section 230 protections in the US enabled social media platforms to profit from harmful content while avoiding accountability. This didn't work then, and won't work now: The worst possible outcome would be to allow generative AI models to continue, whilst also paying licences to large rights holders but not to individual creators, with no meaningful legal recourse to ensure that their work is not ingested into large datasets. This would kill the creativity of the internet.
But there is another way: instead of contorting ourselves to preserve market-based intellectual property in an age of infinite digital abundance, we should embrace the opportunity to create a true digital commons. AI-generated works could lead the way, living in the public domain where anyone can freely use and build upon them - just as we all unknowingly contributed to their creation. Large rights holders should be subject to these free-use models as much as individuals are; perhaps then we shall see the end of the scourge of intellectual property as we currently understand it.
Rather than trying to patch these frameworks, we should use this moment to fundamentally rethink our relationship with knowledge and creation.
As legal scholar Lawrence Lessig argues,
"Codes constitute cyberspaces; spaces enable and disable individuals and groups. The selections about code are therefore in part a selection about who, what, and, most important, what ways of life will be enabled and disabled."
The code embedded in generative AI reveals an important truth - all creation builds on what came before in ways we can't fully trace or credit. What we make becomes part of what others will create. If meaningful authorship is impossible to establish, maybe we should stop trying.
The myth of the solo genius creator was always just that - a myth. Technology has simply made the collaborative, iterative nature of human creativity impossible to ignore. It's time for our legal and economic systems to catch up to that reality.
