For decades, the Copyright Office was a small, sleepy office within the Library of Congress. Each year, her 450 employees at the agency register approximately 500,000 copyrights, ownership rights to creative works, based on a two-century-old law.
But in recent months, the office has suddenly found itself in the spotlight. Lobbyists for Microsoft, Google, the music industry and the news industry are seeking a meeting with copyright registrar Sheila Perlmutter and her staff. Thousands of artists, musicians and technology executives have written to the firm, and hundreds have asked to speak at listening sessions hosted by the firm.
The attention stems from the Copyright Office’s first review of copyright law in the age of artificial intelligence. This technology of harnessing creative content has upended traditional copyright norms, which give owners of books, movies, and music exclusive rights to distribute and copy their works.
This year, the agency plans to release three reports clarifying the government’s position on copyright law related to AI, which will be of great significance and will be used not only in courts but also by legislators and regulators. It will be a very heavy burden.
“We are currently at a very exciting and difficult time as we find ourselves receiving a lot of attention from the broader public,” Perlmutter said.
The technology and media industries debate the value of intellectual property as the Copyright Office reviews it to train new AI models likely to incorporate copyrighted books, news articles, songs, art, and essays. The station was thrust into the middle of a high-stakes conflict between the two countries. Generate text and images. Since the 1790s, copyright law has protected the works of writers and artists so that they can “enjoy the fruits of their intellectual creativity,” the Copyright Office declares on its website.
This law is currently the subject of heated debate. Authors, artists, media companies and others allege that the AI models infringe on their copyrights. The tech companies say they are not copying the material and are using publicly available data on the Internet, which is fair use and within the law. The battle led to lawsuits, including one by the New York Times against ChatGPT’s creators, OpenAI and Microsoft. And copyright holders are calling on authorities to rein in tech companies.
“What the Copyright Office is doing is important because it involves important principles of law and a lot of money,” said Rebecca Tushnet, a professor of copyright and intellectual property law at Harvard Law School. That’s what I did.” “At the end of the day, the question isn’t whether these models exist; it’s who gets paid.”
Congress established the Copyright Office in 1870 to register licenses for books, maps, essays, and other creative works and to deposit those works in the Library of Congress for use by members of Congress. The first children’s book to be registered was the Philadelphia Spelling Book.
When Ms. Perlmutter, a veteran copyright official and former Time Warner intellectual property attorney, was appointed to head the Copyright Office in late 2020, she focused on big technology trends and modernized the office. I promised to lead you to She draws inspiration from past leaders who worked on technological innovations such as cameras, records, Xerox machines, the Internet, and streaming music, and considers how copyright applies in all of them. , the office was required to advise Congress on proposed updates to the law. .
AI has become a hot topic right away. Computer scientist Stephen Thaler sought to copyright an AI-generated work of art by filing an application on his website at the Copyright Office. In 2019, the Bureau rejected his first attempt to register the work. The work, a pixelated scene of railroad tracks running through a tunnel overgrown with bushes and flowers, was dubbed “The Gateway to Modern Paradise.” In February 2022, Ms. Perlmutter rejected a second attempt to register the work for the same reason. Copyright was granted only to original works created by humans.
This decision was the first for works produced by AI and set an important precedent. Perlmutter’s office has been flooded with emails and phone calls from artists and lawmakers, asking him to intervene on how AI companies are also using copyrighted material to train their systems.
In August, she began a formal review of AI and copyright law. The agency said it will investigate whether the use of intellectual property to train AI models violates the law and will further examine whether machine-generated works can be subject to copyright protection. The agency also said it would investigate how AI tools were creating content using individuals’ names, images and likenesses without their consent or compensation.
“There’s a lot of interest in AI,” Perlmutter said in an interview. “Current generative AI systems raise a number of complex copyright issues, some would call existential questions, but they do not require us to address fundamental questions about the nature and value of human creativity. We really need to get started.”
The Secretariat’s interest in the examination was extremely high. The office solicited public comment on the topic and received more than 10,000 responses to a form on its website. The department said he can only receive up to 20 comments during a typical policy review.
In comments on its website, the tech company argued that the way its model incorporates creative content is innovative and legitimate. Andreessen Horowitz, a venture capital firm that has made several investments in AI startups, said in a comment that if AI companies’ content consumption slows, they will “negate at least a decade’s worth of investments, which was the premise of current understanding.” It will jeopardize well-founded expectations.” About the scope of copyright protection in this country. ”
OpenAI, Microsoft, Meta (Facebook’s parent company), and Google are currently relying on a 2015 court ruling in a lawsuit brought by the Authors Guild.
The guild sued Google in 2005 for scanning books for use in excerpts from search engine results and for sharing with libraries. The court ruled that Google did not violate copyright law. Google said scans of the entire book were allowed because it had not made the entire book publicly available, calling it a “transformative” use of copyrighted material. Google relied on an exemption from copyright law known as “fair use,” which allows limited reproduction of copyrighted material for purposes such as criticism, parody, and other transformative uses.
Google, Meta, and AI startup Anthropic all said in comments to the Copyright Office that AI does not copy information to analyze data and reuse it in creative works, etc. repeated the claim.
Authors, musicians, and the media industry claimed that AI companies were robbing them of their livelihoods by taking their content without permission or paying license fees.
“The lack of consent and compensation in this process is theft,” “Family Ties” actress and author Justine Bateman wrote in a comment to the Copyright Office.
News Corp., which publishes the Wall Street Journal and the New York Post, urged the agency to “never lose sight of the simple truth that protecting content creators is one of the core missions of copyright law.” . (The Times also commented.)
Perlmutter said she and a staff of about 20 copyright attorneys review each comment submitted to her firm.
Still, the office may not offer a clear view that satisfies either tech companies or creative people.
“As technology becomes more and more sophisticated, the challenges become exponentially more difficult and the risks and rewards become exponentially greater,” Perlmutter said.
audio creator sarah diamond.
