Sophisticated AI Systems Have Experts Rethinking Regulating Intellectual Property

The democratization of technology has led to, in some ways, the democratization of intellectual property (IP). 

And artificial intelligence (AI) has supercharged this trend. 

“We are fundamentally rethinking the boundaries between what does the platform, or what does the company take on, and what’s dictated by law [as a result of AI],” Arun Sundararajan, the Harold Price Professor of Entrepreneurship and Technology at New York University’s Stern School of Business, tells PYMNTS during a conversation for the TechREG Talks series

Sundararajan explains that platforms like YouTube, TikTok, Alibaba, Amazon, Airbnb, or car rental services — all of which aggregate supply and demand, provide search and discovery, matching, trust, and risk management services — have led to a redistribution of economic activity.

That redistribution has been, and is increasingly being, powered by AI. 

“It’s really the AI that lets you find the song that you’re looking for from that independent musician on Spotify, or like that obscure book that you didn’t know that you were going to like, but you end up liking. AI is also central to platform trust and risk management,” he adds. 

And the emergence of newer, generative forms of AI has further amplified the impact of AI on the sharing economy and raised important questions about regulation and IP.

“What should the private sector do? What should the state do? What is the reach of self-regulation?” asks Sundararajan, noting that it will vary by sector and area of the law, as well as vary by whether something falls under pre-existing IP frameworks. 

“I have never seen a rethinking of regulation because of digital technology that is happening with the same intensity as I have seen over the last year,” he adds. 

The Impact of Artificial Intelligence on Intellectual Property

Generative AI technologies have sparked a rethinking of regulation and IP laws.

Crowd-based capitalism, closely related to the platform business model, has led to a decentralized creator economy, where AI-powered platforms segment users into communities of shared interest. 

The recent advancements in generative AI have the potential to dramatically alter the status quo, says Sundararajan, explaining that the direction of this change will depend on the regulation and policy surrounding intellectual property. It could either lead to a highly centralized world dominated by AI-generated content or expand the abilities and output of millions of human creators.

“It is really going to depend on which way regulation, especially around intellectual property, ends up going,” he adds. 

That’s because AI potentially empowers anyone to become a creator, but also raises the question of ownership of the output of those AI models.

“In many ways, nobody has really thought about protecting a human creator from perfect replication because it simply wasn’t technologically feasible,” says Sundararajan.

Corralling The Disruptive Nature of AI

As generative AI systems become more sophisticated, issues of ownership arise regarding the creative process, the data used for training, and the output of AI systems — and existing intellectual property laws may fall short in resolving these dilemmas.

“Copyright law today doesn’t generally protect artistic style,” explains Sundararajan. “Until now, generally economists would’ve agreed that’s a good thing.”

But the ability of generative AI to replicate a specific artistic style challenges the current notion of copyright protection, which primarily focuses on specific expressions rather than artistic style. This dilemma requires substantial overhaul and a reexamination of the fundamental tradeoffs underlying intellectual property law.

The intersection of artificial intelligence and intellectual property presents complex challenges that require careful consideration and reform. The disruptive nature of AI in the sharing economy and the emergence of generative AI technologies necessitate a reevaluation of existing laws and regulations. 

The recent litigation surrounding the use of copyrighted material in AI-generated outputs highlights the need for reform, highlights Sundararajan. While existing case law may provide some guidance, it may not adequately address the challenges posed by generative AI. 

The arguments around fair use and the impact on the market value of data used for training AI systems will shape the future of IP regulation, he says. However, it is becoming increasingly clear that broader reform is necessary to address the fundamental changes brought about by AI in the realm of intellectual property.

As society grapples with the implications of AI-generated content, it must strike a balance between protecting incentives for creators, ensuring the value of distribution, and fostering future innovation, says Sundararajan, noting that ongoing discussions and legal battles surrounding AI and intellectual property will shape the future of IP law and its role in the age of AI.