As AI’s capabilities increasingly blur the lines between human originality and machine-generated derivatives, traditional copyright frameworks appear increasingly antiquated. The once-clear demarcations of ownership, originality, and rights are now mired in ambiguity. How might we rewrite today’s copyright regulations to enable a future that respects both human ingenuity and the undeniable contributions of generative AI models?
The C/G Boundary
ChatGPT became available to the public on November 30, 2022. I’ve given that date a name: the Curation/Generation AI Boundary (C/G Boundary). The C/G Boundary marks the transition from a time when AI was predominantly used to curate and organize preexisting content to a period characterized by AI’s capability to autonomously generate content and solutions. September 2021, OpenAI’s “knowledge cutoff month,” is also notable because – since then – it has become more challenging to discern content created solely by humans from that which may have had AI assistance or origination.
We are nearing the end of year one after the C/G Boundary, which we can call the Generative Era (GE). At 1GE, no content can be guaranteed to be original. What will life be like in the year 10 GE?
Echoes of the Generative Era
In a dimly lit room, walls adorned with handwritten notes, sketches, and scribbles, James sits hunched over a wooden desk. The room is a sanctuary of silence, punctuated only by the rhythmic scratch of pen on paper. It was a sound that had become rare, a relic of a bygone era, but in 10 GE, it’s so hard to turn off AI-assistance, sequestering yourself away from technology with a physical pen and paper is the only way to ensure your writing is truly your own.
Outside, the city is alive with the pulse of the future. Data-driven messages are ubiquitous, the content shifting and morphing as people interact with data. A stroll down any avenue is a journey of personal discovery. Messages seamlessly transforming from one to another, each pixel catering to your unique tastes and preferences. It is a dance of data and desire in a world where every song, every image, every word is a blend of hybrid creativity and AI precision.
Ten years into the Generative Era everyone from the youngest child to the eldest grandparent is engrossed in their own data-driven world, lost in a sea of content generated just for them. The allure is undeniable. Every piece of information, every form of entertainment is a mirror reflecting their deepest desires. Unsurprisingly, the chasm between the data elite and the data poor has never been wider.
Yet, one-third of the way through a typical 30 year tech cycle, the past is not forgotten. James, a “certified human” with a passion for the written word, continues to create the “old way.” He is a writer. The generative era, with all its allure, has done nothing but inspire him to create authentically human, original works. But what does that mean in 2033?
The Evolution of Copyright in the Age of AI
Alleged and Obvious Violations Leading to AI Mastery
Data Mining without Consent: To train AI models, vast amounts of data are required. In the early days, most companies mined data from public sources, user-generated content, and even private databases without explicit consent. This led to the allegedly unauthorized use of copyrighted materials – this is the subject of several current lawsuits.
Generative Reproductions: AI’s ability to generate content that closely resembles copyrighted works has blurred the lines between originality and replication. This makes it difficult to determine if a piece is a violation, a unique creation or some new mash-up of both.
AI in Art and Music: Generative AI produces art, music, and literature that closely mirrors human-created content. This raised questions about the originality of AI-generated content and its infringement on existing copyrights.
Deepfakes and Media Manipulation: The rise of deepfakes, where AI replicates voices and images of individuals empowers unauthorized reproductions and alterations of copyrighted content.
Rewriting Copyright for the AI Era
Certified Human Badge: To distinguish between human-made and AI-generated content, a “Certified Human” badge could be introduced. This badge would signify that the work is entirely of human origin, without AI intervention. Everyone’s immediate response to this idea is, how would you ensure an honest certification? It will be very hard (if not impossible) to accomplish. But such a certification/identification would go a long way toward solving both training and usage issues.
Contribution Margin: A system could be established to determine the percentage of AI contribution in a piece of work. If AI’s contribution exceeds a certain threshold, the work would be labeled as “synthetic” or “derivative.” This would help in categorizing content and determining its eligibility for copyright protection.
Dynamic Licensing: Traditional licensing models could evolve to accommodate AI’s role. For instance, if an AI model uses a piece of copyrighted content as a reference, a dynamic license could be issued, which takes into account the degree of AI’s involvement. A mechanism like this would require some sort of online registry (similar to the current copyright registry – but rebuilt for the age of AI).
AI as a Tool, Not a Creator: Copyright laws could be revised to view AI as a tool rather than an independent creator. This means that the end-user or the person guiding the AI would hold the copyright, regardless of AI’s contribution. In the age of AI-copilots (2023 and beyond) some version of this will have to evolve. Nothing created with Office 365 or Google Workspace will be totally original after these office suites are fully deployed.
Open-Source and Shared Models: Encouraging open-source AI models and shared content databases could reduce copyright infringements. These platforms would allow creators to willingly contribute their works for AI training, ensuring ethical data usage.
Financial Implications of the AI Era
Losses
Royalty Erosion: With AI reproducing similar content, royalties for original creators will be diminished. The market will be flooded with similar content, reducing the demand for specific copyrighted works. Amazon just limited the number of eBooks you create on a given day. It’s just too easy to create eBooks now.
Litigation Costs: The blurred lines between AI-generated and human-made content has already led to a surge in copyright infringement cases, this trend will continue into the foreseeable future increasing litigation costs for creators and companies.
Devaluation of Original Works: As AI-generated content becomes commonplace, the perceived value of original, human-made works will decrease in many sectors. This is an aspect of this debate that brings out the biggest feelings and most passionate arguments. The best will profit from generative AI, journeymen and posers are not going to fare as well. It’s harsh, but it’s just history repeating itself.
Gains
Cost-Efficient Content Creation: Businesses will benefit from AI’s ability to produce content at a fraction of the cost, leading to increased profitability. Nothing will stop CEO’s and stakeholders from relentlessly pursuing productivity. It’s the nature of capitalism.
Personalized User Experiences: AI’s hyper-personalization capabilities allow for data-driven, hyper-personalized user experiences, leading to increased consumer engagement and sales. The winners in the upcoming general elections will be the candidates that are the best users of data-driven generative AI marketing tools.
Expansion of Creative Horizons: With AI as a co-creator, artists and creators can (and will) explore new avenues, leading to the birth of novel art forms and genres. This is the most exciting part of the generative AI story.
Your Turn
While the essence of originality and the financial value of human-made content faces challenges, the collaborative synergy between humans and AI will also unlock unprecedented creative and economic potentials. We must strike a balance, ensuring that while AI augments human capabilities, human creativity remains protected and cherished.
How would you craft regulations designed to protect creative works (of every kind) while rewarding creators and rights holders 10 years into the Generative Era? I welcome your thoughts.
Author’s note: This is not a sponsored post. I am the author of this article and it expresses my own opinions. I am not, nor is my company, receiving compensation for it. This work was created with the assistance of various AI models, including but not limited to: GPT-4, Bard, Claude, Midjourney, Stable Diffusion, and others.
ABOUT SHELLY PALMER
Shelly Palmer is the Professor of Advanced Media in Residence at Syracuse University’s S.I. Newhouse School of Public Communications and CEO of The Palmer Group, a consulting practice that helps Fortune 500 companies with technology, media and marketing. Named he covers tech and business for , is a regular commentator on CNN and writes a popular . He's a , and the creator of the popular, free online course, . Follow or visit .
