In this article, we will discuss the copyright lawsuit filed by The New York Times against Microsoft and OpenAI in federal court on December 27. This lawsuit is considered one of the most significant lawsuits filed to date regarding copyright infringement by OpenAI and Microsoft. The newspaper claims that the two companies should be held responsible for “billions of dollars in statutory and actual damages.”
Legal Details of the Lawsuit
The New York Times alleges that thousands of its articles have been copied without permission in the training process of the AI models that support OpenAI and Microsoft applications. The newspaper also claims that these AI applications allow users to infringe copyright by producing texts that quote New York Times articles. The newspaper states that the integration of AI models with web browsing and search tools steals commercial referrals and traffic from its own website. The newspaper also argues that its reputation is harmed when OpenAI’s models generate information and falsely attribute it to The New York Times. The newspaper has provided substantial evidence to support these claims, including an exhibit featuring 100 examples of OpenAI models that produce lengthy excerpts of New York Times articles verbatim when prompted with a sentence or part of the original sentence.
The Lawsuit’s Impact on OpenAI
Some experts see this lawsuit as a potential Napster moment for OpenAI, suggesting it could ultimately lead to the company’s bankruptcy. However, the reality is that this lawsuit is likely to be settled. The fact is that the newspaper was negotiating with OpenAI for a licensing agreement and only filed the lawsuit after these negotiations appeared to reach an impasse (possibly because the newspaper was requesting an amount higher than what OpenAI was willing to pay). OpenAI has signed agreements with the associated press agency to license its content for AI training and has also signed an agreement with Axel Springer for access to its current and archival content. The value of this agreement is estimated to be over $10 million annually. Therefore, both OpenAI and Microsoft have a strong incentive to settle rather than deal with years of legal uncertainty.
Potential Intervention by the U.S. Copyright Office or Congress
It is also possible that the U.S. Copyright Office or Congress may intervene before the Supreme Court gets involved. The Copyright Office has concluded a comment period on the implications of creative AI. The Senate has also recently held hearings on this issue. Congress could intervene and pass new legislation that renders The New York Times’ lawsuit invalid. Some legal scholars have suggested that Congress create a “fair use” law that explicitly grants software companies the right to use copyrighted materials for training AI. Meanwhile, those who sympathize with the rights of owners have proposed that lawmakers require creators to be compensated for any works used to train AI. Congress may also insist that AI companies apply filters to screen any results from their models that match copyrighted materials used in training. There is a precedent for Congress intervening in this way: the Audio Home Recording Act of 1992 exempted digital audio tape sellers from copyright infringement lawsuits. However, it also established licensing fees that all manufacturers and importers of digital audio recording devices must pay to the Copyright Office, which distributes this money as copyright payments to music rights holders. Congress could establish a similar system for licensing and ownership of creative software rights.
The Future
Using Copyrighted Material in AI Training
In the end, even if OpenAI ultimately has to pay licensing fees to creators, it may be able to afford it. The company is currently generating revenue at a rate of $1.6 billion annually, and some insiders predict that this figure could reach $5 billion by the end of 2024. With this kind of revenue, OpenAI can continue. Although copyright infringement lawsuits have negatively impacted Spotify’s profits, the company has ultimately managed to reach a settlement regarding music rights. While these payments have affected Spotify’s earnings and the company struggles to convince equity investors of a compelling growth story, Spotify will not go bankrupt.
Therefore, I do not believe that OpenAI will go bankrupt, but I think that the New York Times lawsuit indicates that the era of freely using copyrighted material to train AI is nearing its end. Legal threats will drive most companies building AI models to license any data they use. For example, there are reports that Apple is currently in discussions to obtain a license for the data it seeks to train its AI models. In the field of image production, artists are also increasingly resorting to a technique called obfuscation that makes it impossible to train AI models on their work without their consent. However, this technique is not yet available in the field of text or music, but researchers are working on it. Many publishers have now taken steps to prevent their websites from being freely scraped by bots. Very soon, the only way for companies to obtain the data they need to train good AI models will be if they pay licensing fees. One way or another, the sun sets on the wild west of AI.
With that, we reach the end of the article.
Source: https://www.aol.com/york-times-copyright-suit-means-173229404.html
.lwrp .lwrp-list-row-container{
display: flex;
justify-content: space-between;
}
.lwrp .lwrp-list-row-container .lwrp-list-item{
width: calc(12% – 20px);
}
.lwrp .lwrp-list-item:not(.lwrp-no-posts-message-item){
}
.lwrp .lwrp-list-item img{
max-width: 100%;
height: auto;
object-fit: cover;
aspect-ratio: 1 / 1;
}
.lwrp .lwrp-list-item.lwrp-empty-list-item{
background: initial !important;
}
.lwrp .lwrp-list-item .lwrp-list-link .lwrp-list-link-title-text,
.lwrp .lwrp-list-item .lwrp-list-no-posts-message{
}@media screen and (max-width: 480px) {
.lwrp.link-whisper-related-posts{
}
.lwrp .lwrp-title{
}.lwrp .lwrp-description{
}
.lwrp .lwrp-list-multi-container{
flex-direction: column;
}
.lwrp .lwrp-list-multi-container ul.lwrp-list{
margin-top: 0px;
margin-bottom: 0px;
padding-top: 0px;
padding-bottom: 0px;
}
.lwrp .lwrp-list-double,
.lwrp .lwrp-list-triple{
width: 100%;
}
.lwrp .lwrp-list-row-container{
justify-content: initial;
flex-direction: column;
}
.lwrp .lwrp-list-row-container .lwrp-list-item{
width: 100%;
}
.lwrp .lwrp-list-item:not(.lwrp-no-posts-message-item){
}
.lwrp .lwrp-list-item .lwrp-list-link .lwrp-list-link-title-text,
.lwrp .lwrp-list-item .lwrp-list-no-posts-message{
};
}
Leave a Reply