2023-07-12 00:10:45
PABLO SCARPELLINI
los angeles
Updated Wednesday, July 12, 2023 – 00:10
Writers Paul Tremblay and Mona Awad and comedian Sarah Silverman are suing OpenAI and Meta for the profit they have made from their texts
Illustration: LUCA MARTN
Where does the OpenAI artificial intelligence program get its content from? when you create a text at a devilish speed? Do you copy them from other existing texts or generate your own by mixing a huge amount of data? To what extent is this reworking a plagiarism of texts that are subject to copyright? These are questions that thousands of users have asked themselves and that could find a precise answer as a result of the lawsuits filed by the writers. Paul Tremblay y Mona Awad and, in a separate case, actress SarahSilverman. The whistleblowers claim that the California-based company used their books without consent to train the ChatGPT program and that, in doing so, they violated copyright laws.
In the case of Sarah Silverman, her lawsuit is also directed against goal (Facebook’s parent group) as well as OpenAI, and alleges that the companies illegally used his The Bedwetter memories to train their Artificial Intelligence language models.
Silverman is represented by the same lawyers who have advised Paul Tremblay and Mona Awad and signs their lawsuit along with their colleagues Christopher Golden and Richard Kadrey. All three claim that they never consented to the use of their copyrighted books as training material for intelligence.
Both Silverman and Tremblay Awad are relevant firms. A native of Aurora, Colorado, Tremblay has published such novels as The cabin at the end of the world and A head full of ghosts and is a best-selling author in the horror, fantasy, and science fiction genres. Awad, a Canadian novelist, has been very well received with works like Bunny and 13 ways of looking at a fat girl. Margaret Atwood said of her that it is hers her literary heir apparent. And Silverman has been a star on television since the 1990s when he started working on Saturday Night Live.
Their lawsuits seek to explore the legality of the program that, according to the authors, swallowed his books and used them to train the ChatGPT, to the point of generating very precise summaries of his novels. In the document filed by Tremblay and Awad in a San Francisco court, it is indicated that the company led by Sam Altman and backed by a million-dollar investment from Microsoft has been unduly enriched by the use of your copyrighted material.
OpenAI has avoided openly and transparently share what kind of data and texts you have used to train ChatGPT, but the company did indicate that it is information extracted from the internet, including Wikipedia, and from a large number of archived books. The application refers in particular to the project gutenberga database open to the public with more than 60,000 titles, whose copyrights have already expired.
Tremblay and Awad are the visible heads of a lawsuit that could benefit all persons or entities domiciled in the United States who own the copyright in any work that has been used as training data for the OpenAI language models. His complaint, however, seems to refer to conduct that is difficult to classify since we are not talking about plagiarism but about another form of more indirect profit. The amount of data available on the internet is gigantic and controlling its origin is becoming increasingly complex. However, from some literary circles the decision to present the two lawsuits against OpenAI has been applauded as a way to protect copyrights in a stronger way.
According to the criteria of The Trust Project
Saber ms
#writers #sue #ChatGPT #books #learn #write