Showing posts with label copyrighted works. Show all posts
Showing posts with label copyrighted works. Show all posts

Thursday, January 9, 2025

Elon Musk says all human data for AI training ‘exhausted’; The Guardian, January 9, 2025

, The Guardian; Elon Musk says all human data for AI training ‘exhausted’

"However, Musk also warned that AI models’ habit of generating “hallucinations” – a term for inaccurate or nonsensical output – was a danger for the synthetic data process.

He said in the livestreamed interview with Mark Penn, the chair of the advertising group Stagwell, that hallucinations had made the process of using artificial material “challenging” because “how do you know if it … hallucinated the answer or it’s a real answer”.

Andrew Duncan, the director of foundational AI at the UK’s Alan Turing Institute, said Musk’s comment tallied with a recent academic paper estimating that publicly available data for AI models could run out as soon as 2026. He added that over-reliance on synthetic data risked “model collapse”, a term referring to the outputs of models deteriorating in quality...

High-quality data, and control over it, is one of the legal battlegrounds in the AI boom. OpenAI admitted last year it would be impossible to create tools such as ChatGPT without access to copyrighted material, while the creative industries and publishers are demanding compensation for use of their output in the model training process."

Wednesday, October 23, 2024

Former OpenAI Researcher Says the Company Broke Copyright Law; The New York Times, October 23, 2024

 , The New York Times; Former OpenAI Researcher Says the Company Broke Copyright Law

"Mr. Balaji believes the threats are more immediate. ChatGPT and other chatbots, he said, are destroying the commercial viability of the individuals, businesses and internet services that created the digital data used to train these A.I. systems.

“This is not a sustainable model for the internet ecosystem as a whole,” he told The Times."

Friday, October 11, 2024

Why The New York Times' lawyers are inspecting OpenAI's code in a secretive room; Business Insider, October 10, 2024

  , Business Insider; Why The New York Times' lawyers are inspecting OpenAI's code in a secretive room

"OpenAI is worth $157 billion largely because of the success of ChatGPT. But to build the chatbot, the company trained its models on vast quantities of text it didn't pay a penny for.

That text includes stories from The New York Times, articles from other publications, and an untold number of copyrighted books.

The examination of the code for ChatGPT, as well as for Microsoft's artificial intelligence models built using OpenAI's technology, is crucial for the copyright infringement lawsuits against the two companies.

Publishers and artists have filed about two dozen major copyright lawsuits against generative AI companies. They are out for blood, demanding a slice of the economic pie that made OpenAI the dominant player in the industry and which pushed Microsoft's valuation beyond $3 trillion. Judges deciding those cases may carve out the legal parameters for how large language models are trained in the US."

Tuesday, July 30, 2024

An academic publisher has struck an AI data deal with Microsoft – without their authors’ knowledge; The Conversation, July 23, 2024

Lecturer in Law, University of New England , The Conversation; ; An academic publisher has struck an AI data deal with Microsoft – without their authors’ knowledge

"In May, a multibillion-dollar UK-based multinational called Informa announced in a trading update that it had signed a deal with Microsoft involving “access to advanced learning content and data, and a partnership to explore AI expert applications”. Informa is the parent company of Taylor & Francis, which publishes a wide range of academic and technical books and journals, so the data in question may include the content of these books and journals.

According to reports published last week, the authors of the content do not appear to have been asked or even informed about the deal. What’s more, they say they had no opportunity to opt out of the deal, and will not see any money from it...

The types of agreements being reached between academic publishers and AI companies have sparked bigger-picture concerns for many academics. Do we want scholarly research to be reduced to content for AI knowledge mining? There are no clear answers about the ethics and morals of such practices."

Thursday, July 25, 2024

A new tool for copyright holders can show if their work is in AI training data; MIT Technology Review, July 25, 2024

, MIT Technology Review; A new tool for copyright holders can show if their work is in AI training data

"Since the beginning of the generative AI boom, content creators have argued that their work has been scraped into AI models without their consent. But until now, it has been difficult to know whether specific text has actually been used in a training data set. 

Now they have a new way to prove it: “copyright traps” developed by a team at Imperial College London, pieces of hidden text that allow writers and publishers to subtly mark their work in order to later detect whether it has been used in AI models or not. The idea is similar to traps that have been used by copyright holders throughout history—strategies like including fake locations on a map or fake words in a dictionary. 

These AI copyright traps tap into one of the biggest fights in AI. A number of publishers and writers are in the middle of litigation against tech companies, claiming their intellectual property has been scraped into AI training data sets without their permission. The New York Times’ ongoing case against OpenAI is probably the most high-profile of these.  

The code to generate and detect traps is currently available on GitHub, but the team also intends to build a tool that allows people to generate and insert copyright traps themselves."

Friday, January 26, 2024

George Carlin Estate Sues Creators of AI-Generated Comedy Special in Key Lawsuit Over Stars’ Likenesses; The Hollywood Reporter, January 25, 2024

 Winston Cho, The Hollywood Reporter ; George Carlin Estate Sues Creators of AI-Generated Comedy Special in Key Lawsuit Over Stars’ Likenesses

"The complaint seeks a court order for immediate removal of the special, as well as unspecified damages. It’s among the first legal actions taken by the estate of a deceased celebrity for unlicensed use of their work and likeness to manufacture a new, AI-generated creation and was filed as Hollywood is sounding the alarm over utilization of AI to impersonate people without consent or compensation...

According to the complaint, the special was created through unauthorized use of Carlin’s copyrighted works.

At the start of the video, it’s explained that the AI program that created the special ingested five decades of Carlin’s original stand-up routines, which are owned by the comedian’s estate, as training materials, “thereby making unauthorized copies” of the copyrighted works...

If signed into law, the proposal, called the No AI Fraud Act, could curb a growing trend of individuals and businesses creating AI-recorded tracks using artists’ voices and deceptive ads in which it appears a performer is endorsing a product. In the absence of a federal right of publicity law, unions and trade groups in Hollywood have been lobbying for legislation requiring individuals’ consent to use their voice and likeness."

Tuesday, July 25, 2023

The Generative AI Battle Has a Fundamental Flaw; Wired, July 25, 2023

  , Wired; The Generative AI Battle Has a Fundamental Flaw

"At the core of these cases, explains Sag, is the same general theory: that LLMs “copied” authors’ protected works. Yet, as Sag explained in testimony to a US Senate subcommittee hearing earlier this month, models like GPT-3.5 and GPT-4 do not “copy” work in the traditional sense. Digest would be a more appropriate verb—digesting training data to carry out their function: predicting the best next word in a sequence. “Rather than thinking of an LLM as copying the training data like a scribe in a monastery,” Sag said in his Senate testimony, “it makes more sense to think of it as learning from the training data like a student.”...

Ultimately, though, the technology is not going away, and copyright can only remedy some of its consequences. As Stephanie Bell, a research fellow at the nonprofit Partnership on AI, notes, setting a precedent where creative works can be treated like uncredited data is “very concerning.” To fully address a problem like this, the regulations AI needs aren't yet on the books."

Wednesday, January 18, 2023

AI Trained on Copyrighted Works: When Is It Fair Use?; Lexology, January 16, 2023

Diana Bikbaeva - Diana Bikbaeva, Lexology; AI Trained on Copyrighted Works: When Is It Fair Use?

"We recently published an article that got much traction about whether machine learning on copyrighted materials is fair use. We now offer a deep analysis with new ideas and suggestions on legal risk mitigation and principles for more ethical AI systems.

AI is still a relatively new, although rapidly evolving technology, and some of its legal implications (especially in copyright law) remain a gray area, creating uncertainty on its use and development.

AI and machine learning technology are not one-size-fits-all and have diverse structures and algorithms specific to the tasks they are programmed to solve. So, any discussion of the legal implications of machine learning and resulting artificial intelligence needs to avoid sweeping conclusions on the technology in general and should consider the underlying technology and its treatment of copyrighted materials on a case-by-case basis."

Wednesday, December 21, 2022

AI-Created Comic Has Been Deemed Ineligible for Copyright Protection; CBR, December 20, 2022

BRIAN CRONIN, CBR; AI-Created Comic Has Been Deemed Ineligible for Copyright Protection

"The United States Copyright Office (USCO) reversed an earlier decision to grant a copyright to a comic book that was created using "A.I. art," and announced that the copyright protection on the comic book will be revoked, stating that copyrighted works must be created by humans to gain official copyright protection. 

In September, Kris Kashtanova announced that they had received a U.S. copyright on his comic book, Zarya of the Dawn, a comic book inspired by their late grandmother that she created with the text-to-image engine Midjourney. Kashtanova referred to herself as a "prompt engineer" and explained at the time that she went to get the copyright so that she could “make a case that we do own copyright when we make something using AI.”"

Tuesday, May 3, 2022

The Seizure of Jewish Intellectual Property Ahead of World War II; Library of Congress, April 28, 2022

, Library of Congress; The Seizure of Jewish Intellectual Property Ahead of World War II

"The following is a guest post by Marilyn Creswell, information resources assistant at the University of Michigan Law School. She served as Librarian-in-Residence at the U.S. Copyright Office from July 2020 to April 2021.

As the United States enters the Days of Remembrance of the Victims of the Holocaust, we remember the many hardships Jewish people have overcome. In this blog we specifically explore the lesser-known area of intellectual property (IP) leading up to and during World War II. Beginning in 1933, the Nazi German state began pressuring Jewish business owners to sell their businesses far below market value. By 1938, a majority of Jewish-owned businesses were already sold or out of business when this process, called Aryanization, became compulsory after Kristallnacht.1 As part of the seizure of businesses and personal property, the ability of Jewish people to benefit from their intellectual property was also severely restricted. A 1939 executive order required all Jewish men to add “Israel” as a second name and women to add “Sara.”2 This made it easier for Nazi officials to deny intellectual property registrations and renewals to Jewish applicants, cutting them off from the IP system.3 While the loss of IP rights pales in comparison to the horrific death tolls during World War II, its loss is another indignity the Jewish people suffered and source of wealth extracted at the hands of the Nazis.

In some instances, works by Jewish authors were nearly completely reproduced and distributed by others without their consent. One example of an Aryanized work is Alice Urbach’s So kocht man in Wien!, a Viennese cookbook. Urbach was forced to transfer the rights to her book, which was then republished with new authorial credit to “Rudolf Rösch.” The new work kept most of the original texts and photographs of her cooking demonstrations but removed elements celebrating Vienna’s diversity.4 In the field of medicine, Dr. Josef Löbel’s Knaurs Gesundheitslexikon was a health encyclopedia that, after the Otto Liebmann publishing house was taken over by a Nazi publisher, was republished by the author Herbert Volkmann under the pseudonym “Peter Hiron.” Volkmann even added new sections on race, homosexuality, and prison psychology. He similarly usurped authorship for Dr. Walter Guttman’s Medizinische Terminologie and its ongoing publications.5

Public domain works were revised to remove references to Jewish people and culture. For example, Fritz Stein presented a new version of Handel’s Occasional Oratorio (Gelegenheits Oratorium) in 1935 that added state-promoting verses and removed references to Jacob, Jehovah, and the full aria “When Israel, like the bounteous Nile.” In 1941, Handel’s Jephtha was renamed Das Opfer and changed so its Jewish history was reframed as a broader narrative about nationalism. The text of his Judas Maccabeus was not only rewritten to omit Jewish references, but it went so far as to make it into a “patriotic fold oratorio” and eventually transplanted Judas with a Field Marshall, a powerful military dictator analogous to the Führer.6 Also in 1941, all theatrical productions required permission from the Reich Dramaturgy, which banned Shakespeare’s historical plays but encouraged the broadcast and production of the anti-Semitic Merchant of Venice.7"

Thursday, May 27, 2021

Ode to Cicadas; Library of Congress, May 24, 2021

,  Library of Congress; Ode to Cicadas

"They’re flying, buzzing, and crawling everywhere! Washington, DC, neighborhoods around the U.S. Copyright Office are teeming with Brood X cicadas, taking their next steps on a seventeen-year journey. Along the way, they’re also inspiring musicians, photographers, artists, and authors to create copyrighted works."

Tuesday, January 21, 2020

A Tool That Removes Copyrighted Works Is Not a Substitute for Fair Use; Electronic Frontier Foundation (EFF), January 20, 2020

Katharine Trendacosta, Electronic Frontier Foundation (EFF);

A Tool That Removes Copyrighted Works Is Not a Substitute for Fair Use


"By making eliminating material flagged by Content ID so easy—just click here!—and making challenging matches so perilous, YouTube has put its thumb on the scale against fair use and in favor of copyright abuse. That thumb gets especially heavy given how few real alternatives to YouTube exist.

Hosting creative content should mean a robust commitment to fair use. Fair use enriches our culture and our understanding of it. It is what ensures that copyright doesn’t strangle free expression and creativity. Subtle reinforcement of anti-fair use ideas enacted by private companies, done by the largest players in the ecosystem, does real damage."

Saturday, December 29, 2018

New Life for Old Classics, as Their Copyrights Run Out; The New York Times, December 29, 2018

Alexandra Alter, The New York Times; New Life for Old Classics, as Their Copyrights Run Out

"This coming year marks the first time in two decades that a large body of copyrighted works will lose their protected status — a shift that will have profound consequences for publishers and literary estates, which stand to lose both money and creative control.

But it will also be a boon for readers, who will have more editions to choose from, and for writers and other artists who can create new works based on classic stories without getting hit with an intellectual property lawsuit...

Friday, December 21, 2018

For the First Time in More Than 20 Years, Copyrighted Works Will Enter the Public Domain; Smithsonian Magazine, January 2019

, Smithsonian Magazine;
 
"At midnight on New Year’s Eve, all works first published in the United States in 1923 will enter the public domain. It has been 21 years since the last mass expiration of copyright in the U.S."

Thursday, June 7, 2018

Copyright Term: When Does a Work Enter the Public Domain?; Lexology, May 24, 2018

Workman Nydegger - J. Dustin Howell, Lexology; Copyright Term: When Does a Work Enter the Public Domain?

"Years ago, there was a radio show that featured a lawyer who would take calls from listeners and answer their legal questions. A man called into the show and said he had obtained some photographs of warplanes from a museum. A cable network had contacted the man about using the photographs as part of a documentary it was making. This man was wise enough to know that the photographs may very well be protected by copyright. He wanted to know how to find out if the photographs were still protected by copyright and who owned them, because without owning the copyrights himself or having a license from the copyright owner, the man could not authorize the cable network to use the photographs. Whether a work (e.g., photograph, literature, music, sculpture, movie, etc.) is protected by copyright or whether it is in the public domain for anyone to use can depend on a variety of factors."

Thursday, January 5, 2017

Bulgarians Listen to Classics Thanks to Copyright War; Balkan Insight, 1/5/17

Mariya Cheresheva, Balkan Insight; Bulgarians Listen to Classics Thanks to Copyright War

"At 00.01 am on January 1, 2017, instead of hearing the official Bulgarian anthem, as they do every year, listeners to Bulgarian National Radio, BNR, were surprised to hear an alternative version performed by BNR’s own choir and symphonic orchestra.


This was not an independent decision of the music editors of BNR. It turned out that they had been banned from playing the official national anthem owing to a decision of Musicautor, Bulgaria’s non-profit society of composers, lyricists and music publishers, which exists to collectively manage copyright issues.

Musicautor, which hold the copyright to over 14,000,000 songs of Bulgarian and worldwide artists, suspended its contract with BNR from the beginning of the new year, demanding higher fees.

It has banned BNR from playing much contemporary Bulgarian and foreign music until the fee issue is resolved."

Friday, May 20, 2016

Avengers and X-Men wage war in action-packed fan trailer; Comic Book Resources, 5/20/16

Kevin Melrose, Comic Book Resources; Avengers and X-Men wage war in action-packed fan trailer:
"Fans at last saw Spider-Man introduced into the Marvel Cinematic Universe with Captain America: Civil War, but it will be a long, long time before they witness the X-Men do the same. Until then, they can watch this epic fan trailer, which pits Marvel’s mutants against Earth’s Mightiest Heroes, over and over and over again."