Think of your favorite song, movie or music. Then think about how you enjoy, watch, listen or engage with that content. The answer? Copyright.
Copyright has not disappeared in the age of AI, but it is being stress-tested in new ways. For example, think about music. Digital service providers (DSPs) now report tens of thousands AI generated tracks uploaded daily. Some of these tracks even rose to the top of the listening charts, competing with works created by humans. Production systems now produce text, visual art and music at scale, blurring the lines between inspiration, simulation and automation. Against this backdrop, the current debate about copyright will determine whether it remains a living promise to creators or becomes an artifact that companies politely accept but can quietly navigate.
I have spent my career defending copyright on behalf of record labels, studios and creators. Now I teach business students who are serious about both building with AI and protecting the sanctity of human intelligence. If ever copyright was at a crossroads, that time is now. So much time to educate the future leaders of the entertainment industry about something that is changing before our eyes.
A work designed for the emergence of copyright
Before joining the full-time faculty of American University’s Kogod School of Business, where I was the director of The billboard is noticed Business and Entertainment Program, I have spent decades working for companies that rely on patents—both protecting and innovating; Copyright not only protects catalogs of creative works, but also allows investment in new artists, stories and songs. Without predictable ownership and licensing structures, there is little incentive to finance risk.
In 2013, I left corporate America out of concern that individual songwriters, filmmakers and other independent creators would be drowned out by better-funded voices when writing policy lines. By 2026, AI technology has proven many of those concerns have been put to rest
Unlike the new distribution model, AI systems can import multiple patented operations in seconds, produce output at scale and compete directly with what you’ve trained. At this point, the rise of AI solidifies the premise that the contribution of the human writer should be the anchor of the creative ecosystem.
What the law says is silent
Through all this turmoil and uncertainty, US law so far has sent a clear message: people are at the center of copyright. The US Copyright Office has emphasized that copyright is necessary for protection; Only mechanical works are not eligible, a position the courts have confirmed in similar cases Thaler v. Perlmutter.
At the same time, courts and agencies argue that using large amounts of copyrighted material to train AI models amounts to infringement or “fair use.” A growing wave of lawsuits—filed by writers, musicians, music publishers and media companies—produced early rulings that cut across the spectrum. Some judges emphasized flexible use, while others indicated restrictions.
While we wait for these court opinions to help fix traffic laws, some rights holders are turning to license agreements and indemnities. These private discussions between AI companies and publishers, music rights organizations and media companies indicate that this use of training will be given an express license rather than just being exploited in the future, an impact that no one can be sure of yet. These agreements can help legitimize compensation structures, but they also raise questions about who has bargaining power and who doesn’t. Large catalog owners may secure deals. Independent creators often don’t have the same bargaining power. That inequality represents the current tension.
Validity of AI model training
When readers ask if it’s “legal” to train AI models on copyrighted works released online, they’re asking a deeper question: “Is my work worth it if a machine can learn from it for free?” Today’s business students are digital natives and emerging franchisees. Many are also artists themselves. They are experimenting with AI tools for documentation, design campaigns and prototype businesses. And importantly, I’m not telling them to turn their backs on this technology. They look at how and if work in the entertainment industry will sustain in the coming years. I expect the presence of AI to continue to grow, so they need to understand it deeply in order to have an impact on their fields of interest.
Addressing basic concerns about the value of their work is easy, but the answer to the legal question is complex and complicated. The current copyright framework is not designed for a world where an AI system can absorb millions of books or songs in an instant, and produce output at scales that feel close enough to the real thing to reduce their value. As technology advances, so does the law.
Rather than offering simple answers, I encourage readers to consider three questions that increasingly shape responsible engagement:
- Who owns the input to the model you are using, and have they actually chosen to provide content to the model?
- Does your output add to someone’s creativity, or does it simply replace someone else’s work?
- If it succeeds, will the creators whose work helped train these programs share in that success, or will they be marginalized again?
Those discussions are not theoretical. They will shape the way this next generation writes contracts, builds companies and shapes policy debates. If we do this right, AI learning and copyright knowledge will be taught in parallel.
So, can copyright exist in the age of AI?
The current era is marked by overwhelming pressures: ongoing litigation, regulatory scrutiny, secret licensing deals and employee concerns across the creative industries. At the same time, AI adoption is accelerating within studios, agencies and record labels. Companies try as the courts deliberately do.
The future of copyright in the age of AI depends on whether policymakers, companies and educators affirm its primary purpose: to protect and promote human creativity. Maintaining a balance between innovation and creativity will require deliberate choices about consent, compensation and accountability before norms are properly calculated.
I have spent my career trying to make sure that creators are not treated as collateral damage to someone else’s creative story. In the age of AI, copyright can still be a living promise, but only if we remember who it was written for—and if we give those human voices real power in shaping what comes next.
