AI expert Marva Bailer gives her opinion on Sarah Silverman and other authors' lawsuit against OpenAI and why it's not so much about money.
Sarah Silverman and other authors' lawsuit against OpenAI, the company behind ChatGPT, has stalled.
The original complaint from Silverman, Richard Kadrey, Christopher Golden and Paul Tremblay alleges that OpenAI used their books without permission to develop its so-called large language models, which their makers pitch as powerful tools for automating tasks by replicating human conversation .
On Monday, a federal judge dismissed several claims in the lawsuit, including claims of copyright infringement, negligence and unjust enrichment.
AI expert Marva Bailer told FOX Business that we can expect more cases like this in the future.
SARAH SILVERMAN, AUTHORS SUING META, OPENAI FOR ALLEGED COPYRIGHT INFRINGEMENT
Sarah Silverman and other authors suffered a setback in their lawsuit against OpenAI after a federal judge dismissed several of the plaintiffs' claims. (Mark Von Holden/Variety via Getty Images / Getty Images)
“The verdict in this case will be consistent with many other cases,” she said. “They are trying to prove that the input to OpenAI, a large language model, will be a derivative of their original work. And what determines that is the output of the work. So that's what this case means. You are.” Examining the edition to determine whether copyright exists on the likeness of the original work.
WATCH: AI expert weighs in on Sarah Silverman's OpenAI lawsuit
AI expert Marva Bailer gives her opinion on Sarah Silverman and other authors' complaint against OpenAI and why it's not so much about money.
The copyright infringement lawsuit states that the responses generated by OpenAI's ChatGPT program are an infringing work that is only possible through information obtained from copyrighted material, in this case Silverman's book “The Bedwetter” and works by other authors like Paul Tremblay’s “The Cabin.” at the end of the world.”
“Plaintiffs do not explain what the results contain or claim that any particular edition is substantially or at all similar to their books. “Accordingly, the court dismisses the copyright infringement claim with leave to modify,” U.S. District Judge Araceli Martinez-Olguin wrote an order.
“This trial will be one of many,” Bailer said. “And for artists to win and writers to win, they have to connect the dots. And this will be a very challenging opportunity for people to see these connections.”
WHAT IS ARTIFICIAL INTELLIGENCE (AI)?
A photo illustration of the ChatGPT and OpenAI logo on a phone held by a person. (CFOTO/Future Publishing via Getty Images / Getty Images)
A claim under the Digital Millennium Copyright Act that “restricts the removal or modification of copyright management information.” [also known as CMI]“such as title, author and copyright holder were also rejected.
Judge Martinez-Olguin sided with OpenAI and wrote: “Plaintiffs provide no facts to support this claim. In fact, the complaints contain excerpts from ChatGPT output that contain several references to the plaintiffs' names, suggesting that OpenAI has not removed all references to “the name of the author.”
WATCH: AI expert drops Sarah Silverman and others' lawsuit against OpenAI
AI expert Marva Bailer explains what Sarah Silverman and other authors' lawsuit against OpenAI is about
“Even if plaintiffs presented facts showing that defendants knowingly removed CMI from the books during the training process, plaintiffs have not shown how the omission of CMI from the copies used in the training set gave defendants reasonable grounds to believe.” stated that the issuance of ChatGPT would result in, enable, facilitate or conceal violations of the law.”
Bailer noted, “What we’re looking at here is the Fair Use Act. So one of the arguments is: Is this education? And if it is education then it would be fine to use these works. But where are the authors and? The challenge for creators is that they feel like they are losing out on compensation, but they need to understand the context. And this is where the nuances come into play.”
CLICK HERE TO READ MORE ABOUT FOX BUSINESS
AI expert Marva Bailer noted that Silverman and other Hollywood stars need to “see the connections” to prove they will lose compensation in such cases. (Jason Kempin/Getty Images / Getty Images)
The claims of creating and distributing “derivative works” were also denied. Silverman and the other plaintiffs originally claimed that “any edition of the OpenAI Language Models is an infringing derivative work.”
However, the judge ruled that this had been done “without specifying what such results entail – that is, whether they are copyrighted books or copies of the books”, calling this “insufficient” to satisfy their claim underpinning the Digital Millennium Copyright Act.
Negligence and unjust enrichment claims were dismissed due to a lack of proof of legal obligation and relationship between the parties and no allegation that OpenAI obtained “unjustified” benefits from the copyrighted works through fraud or other means.
Upholding the unfair competition claim, the judge wrote: “Assuming that Plaintiffs' allegations are true – that Defendants have used Plaintiffs' copyrighted works to train their language models for commercial purposes – the Court concludes that.” Conclude that the defendant’s conduct may constitute an unfair practice.”, this part of the UCL [The Unfair Competition Law of California] The claim can be continued.
Promising new technologies have a 'staggeringly difficult' copyright problem: expert
The judge upheld Silverman and her co-plaintiffs' claim of unfair competition. (Kevin Mazur/Getty Images / Getty Images)
Silverman and her co-plaintiffs have until March 13 to file amendments to their complaint and move forward with this lawsuit.
FOX Business reached out to both Silverman and OpenAI but did not immediately respond to requests for comment.
“The plaintiffs do have the opportunity to come back and clarify the facts. But this looks like they have to prove that they have either lost income or lost the opportunity for their brand,” Bailer explained.
Bailer noted that could be difficult for a Hollywood star like Silverman and best-selling authors.
GET FOX BUSINESS ON THE GO by CLICKING HERE
Bailer pointed out that it could be difficult for Hollywood stars to prove that they were harmed by companies like OpenAI. (Emma McIntyre/Getty Images for Academy Museum of Motion Pictures / Getty Images)
“In many cases it's going to be very, very hard to prove that because not only do they have books, they've just released podcasts, they've got films, they've got Broadway plays. So to be able to tie everything back to that, this open AI language model is going to be a big challenge.”
In addition, in Bailer's opinion, the lawsuit is less about economic aspects and more about intellectual aspects.
“I don’t think this argument is about compensation. It’s about control and consistency,” she said. “And this is a new model. So when you see your name in places that you didn't know it was going to be, and by the way, it might not be exactly what you wanted to say, or the connection is going to challenge you, but it might also open up a whole new audience or experience you've never thought of before. So it will require a lot of patience.
“They think about the then and now and try to protect the brand and their image, but they don’t think about the future economies that they could really be a part of.”