A new lawsuit may bring AI copyright law into clearer focus, continuing a common debate on the ethics of machine learning. Authors Paul Tremblay (“The Cabin at the End of the World”) and Mona Awad (“Bunny” and “13 Ways of Looking at a Fat Girl”) claim OpenAI used their books to train ChatGPT without their permission, thereby infringing copyright.
ChatGPT produces text by calculating the probability of the next best word based on an unimaginably large corpus of data. The controversy arises from the authors’ assertion that their works’ precise summaries required direct training on their books. Even if this turns out to be true (which will be either very easy or extremely hard to prove), it is unclear whether reading something and understanding it (and being able to summarize it for profit) is a violation of copyright law.
Regardless of the verdict, this case underscores the urgent need for explicit norms in AI ethics and copyright law, paving the way for a future where AI developers must tread transparently and respectfully around their data sources and training techniques.
Just for fun, I asked ChatGPT Plus (in default mode – no internet access) the following:
Can you summarize the book “Television Disrupted: The transition from network to networked TV” by Shelly Palmer (Focal Press, 2006)?
ChatGPT
I don’t have the ability to summarize “Television Disrupted: The Transition from Network to Networked TV” by Shelly Palmer (Focal Press, 2006) as I can’t directly access or read books. However, I can tell you that Shelly Palmer is a well-regarded technology and media analyst, and the book’s title suggests that it would cover the significant changes in the television industry caused by the advent of the internet and digital technologies.
Some themes that are likely covered in the book, based on Palmer’s expertise and the title, might include:
The transition from traditional broadcasting to the internet: The ways that television distribution has changed due to streaming and online platforms.
Changes in media consumption: How viewers’ habits have evolved due to the ability to watch content on-demand, binge-watch series, and access a much wider range of content than was available through traditional broadcasting.
The impact on the television industry: The effects of these changes on broadcasters, advertisers, and others involved in the industry, including the shift in advertising revenues and the rise of new players like Netflix, Amazon, and YouTube.
The future of television: Predictions about how these trends might continue to evolve in the future, including the increasing convergence of television and other forms of digital media, potential technological advancements, and the ever-changing landscape of the media industry.
For a detailed summary or analysis of this specific book, I recommend reading it directly or looking for reviews or summaries online that were written by people who have read the book.
This summary is absolutely accurate, as if it had read the book. I repeated this query several times in several different chat windows to ensure that it gave consistent answers. It did. Each was slightly different, but if it didn’t explicitly tell me that it had not read “Television Disrupted,” I would not believe it didn’t.
This lawsuit is going to be fascinating. If you want to gain a deeper understanding of the problem, sign up for my free online course Generative AI for Execs. It will help you build compelling arguments for both sides of the issue.
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 LinkedIn’s “Top Voice in Technology,” he covers tech and business for Good Day New York, is a regular commentator on CNN and writes a popular daily business blog. He's a bestselling author, and the creator of the popular, free online course, Generative AI for Execs. Follow @shellypalmer or visit shellypalmer.com.
