U.S. Judge Allows Artists' Copyright Lawsuit Against AI Companies to Proceed

Big win for artists' AI copyright lawsuit in the US

Kelly McKernan, one of ten artists who sued tech companies using Stable Diffusion to generate AI images, hailed this week's ruling as 'very exciting'. A copyright lawsuit filed by a group of artists against companies using text-to-image generators such as Stability AI, Midjourney and DeviantArt has been allowed to proceed by a US judge. In a ruling this week, US district judge William Orrick said that motions brought forward by artists Sarah Andersen, Kelly McKernan, Karla Ortiz, Hawke Southworth, Grzegorz Rutkowski, Gregory Manchess, Gerald Brom, Jingna Zhang, Julia Kaye and Adam Ellis were granted in part and dismissed in part. The class action lawsuit alleged that Stable Diffusion, the AI image generator created by start-up Stability AI, used the artists' works as "training images" to produce AI-generated images "in the style" of the original images. Stability AI's text-to-image generator, Stable Diffusion, can create new images based on a sentence a user inputs, combining different concepts, attributes and styles. These AI models like Midjourney and DALL-E are trained using a massive number of existing images. However, concerns have been raised that some of the content used to train these models could be copyright protected and taken without the consent of artists. The lawsuit also extended to Midjourney, DeviantArt and Runway AI, all of whom allegedly used Stable Diffusion to generate images that the artists believe violated their copyright. 'Long road ahead' McKernan, who is based in Nashville, Tennessee, hailed the ruling as "very exciting". "The judge is allowing our copyright claims through and now we get to find out all the things these companies don't want us to know in discovery. This is a HUGE win for us. I'm SO proud of our incredible team of lawyers and fellow plaintiffs!" she wrote on X. "We have a long road ahead still but that doesn't mean we shouldn't acknowledge all the effort it took to get to this milestone in our lawsuit. It's been an intense 18 months, let's keep going!" McKernan initially sued the companies in January 2023 along with Andersen and Ortiz, one of several prominent copyright lawsuits against AI companies. While Orrick dismissed many of their allegations in October, he allowed them to be refiled. Last February, Getty Images sued Stability AI in a US court, accusing the start-up of "brazen infringement" of its intellectual property "on a staggering scale". It allegedly copied more than 12m photos - along with captions and metadata - from Getty Images' collection without obtaining permission or compensating them. The move marked an escalation over the previous month, when Getty Images sued Stability AI in London for the same reasons, claiming it "unlawfully copied and processed" millions of copyright-protected images for its own commercial benefit and "to the detriment of content creators". In November, former vice-president of audio at Stability AI Ed Newton-Rex resigned from his post after disagreeing with the company's opinion that training generative AI models on copyrighted works is "fair use". He said that despite his colleagues having a more "nuanced" view on this issue than some competitors, he wasn't able to change the "prevailing opinion" on fair use at the company. Find out how emerging tech trends are transforming tomorrow with our new podcast, Future Human: The Series. Listen now on Spotify, on Apple or wherever you get your podcasts.

Judge rules that artists copyright claims against AI companies can proceed - SiliconANGLE

Judge rules that artists copyright claims against AI companies can proceed A U.S. District Court judge has given permission to a group of artists to pursue claims against several generative artificial intelligence image and video-making companies that alleges that their services infringe on their copyrights. The case was brought by three artists against Stability AI Ltd., DeviantArt Inc. and Midjourney Inc. in January 2023 and alleged that the companies make unauthorized use of existing copyright images. A generative AI model is trained based on existing works to create new works, with the artists claiming that the training of the AI models, allegedly based on their works - it has not been proven that the models were - constitutions copyright theft. A subsequent court ruling in October found that the copyright infringement claims against Midjourney and DeviantArt by plaintiffs Kelly McKernan, Karla Ortiz and Sarah Andersen were "defective in numerous respects" and could not sustain themselves. However, the court at the time found that a single direct claim alleging direct infringement against Stability AI could proceed. Reuters reports that U.S. District Judge William Orrick has now ruled that the artists behind the court case have reasonably argued that the companies violate their rights by illegally storing their works on their systems. The judge did, however, throw out other claims that accused the AI companies of unjust enrichment, breach of contract and breaking a separate U.S. copyright law. Notably, the decision did not address the core claim that the accused companies misused the artist's works to train their AI systems directly and hence infringed on copyright. Nor did the judge rule on whether the use of such works constitutes fair use under Section 107 of the U.S. Copyright Act. An attorney representing the artists said in a statement that the ruling was "a significant step forward for the case," despite the fact that the judge did not rule on the core copyright components alleged by the artists. AI has exploded onto the scene through the first half of the third decade of the 21st century and when historians look back at this time, it will be the decade that AI first came to the forefront as a potentially world-changing technology. But as history shows, new technology always has its luddites who stand against progress. The artists, in this case, are clearly the 21st-century version of the 19th-century cotton and wool mill workers who stood against the progress of mechanical and automated machinery. While it's easy to sympathize with the artists who are being replaced by AI, progress is what it is. Even if they were to win their case in the U.S. and, by some miracle, AI companies were not able to continue in the country, there are 194 other countries in the world that are not subject to U.S. court rulings - progress in AI will continue irrespective of three artists suing a handful of AI companies.

Artists celebrate as copyright infringement case against AI image generators moves forward

Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More Visual artists who joined together in a class action lawsuit against some of the most popular AI image and video generation companies are celebrating today after a judge ruled their copyright infringement case against the AI companies can move forward toward discovery. Disclosure: VentureBeat regularly uses AI art generators to create article artwork, including some named in this case. Which artists are involved? Artists Sarah Andersen, Kelly McKernan, Karla Ortiz, Hawke Southworth, Grzegorz Rutkowski, Gregory Manchess, Gerald Brom, Jingna Zhang, Julia Kaye, and Adam Ellis have, on behalf of all artists, accused Midjourney, Runway, Stability AI, and DeviantArt of copying their work by offering AI image generator products based on the open source Stable Diffusion AI model, which Runway and Stability AI collaborated on and which the artists alleged was trained on their copyrighted works in violation of the law. What the judge ruled today While Judge William H. Orrick of the Northern District Court of California, which oversees San Francisco and the heart of the generative AI boom, didn't yet rule on the merits of the case, he wrote in his decision issued today that the "the allegations of induced infringement are sufficient," for the case to move forward toward a discovery phase -- which could allow the lawyers for the artists to peer inside and examine documents from within the AI image generator companies, revealing to the world more details about their training datasets, mechanisms, and inner workings. "This is a case where plaintiffs allege that Stable Diffusion is built to a significant extent on copyrighted works and that the way the product operates necessarily invokes copies or protected elements of those works," Orrick's decision states. "Whether true and whether the result of a glitch (as Stability contends) or by design (plaintiffs' contention) will be tested at a later date. The allegations of induced infringement are sufficient." Artists react with applause "The judge is allowing our copyright claims through & now we get to find out allll the things these companies don't want us to know in Discovery," wrote one of the artists filing the suit, Kelly McKernan, on her account on the social network X. "This is a HUGE win for us. I'm SO proud of our incredible team of lawyers and fellow plaintiffs!" "Not only do we proceed on our copyright claims, this order also means companies who utilize SD [Stable Diffusion] models for and/or LAION like datasets could now be liable for copyright infringement violations, amongst other violations," wrote another plaintiff artist in the case, Karla Ortiz, on her X account. Technical and legal background Stable Diffusion was allegedly trained on LAION-5B, a dataset of more than 5 billion images scraped from across the web by researchers and posted online back in 2022. However, as the case itself notes, that database only contained URLs or links to the images and text descriptions, meaning that the AI companies would have had to separately go and scrape or screenshot copies of the images to train Stable Diffusion or other derivative AI model products. The case, recorded under the number 3:23-cv-00201-WHO, was originally filed back in January of 2023. It has since been amended several times and parts of it struck down, including today. A silver lining for the AI companies? Orrick did hand the AI image generator companies a victory by denying and tossing out with prejudice claims filed against them by the artists under the Digital Millennium Copyright Act of 1998, which prohibits companies from offering products designed to circumvent controls on copyrighted materials offered online and through software (also known as "digital rights management" or DRM). Midjourney tried to reference older court cases "addressing jewelry, wooden cutouts, and keychains" which found that resemblances between different jewelry products and those of prior artists could not constitute copyright infringement because they were "functional" elements, that is, necessary in order to display certain features or elements of real life or that the artist was trying to produce, regardless of their similarity to prior works. The artists claimed that "Stable Diffusion models use 'CLIP-guided diffusion" that relies on prompts including artists' names to generate an image. CLIP, an acronym for "Contrastive Language-Image Pre-training," is a neural network and AI training technique developed by OpenAI back in 2021, more than a year before ChatGPT was unleashed on the world, which can identify objects in images and label them with natural language text captions -- greatly aiding in compiling a dataset for training a new AI model such as Stable Diffusion. "The CLIP model, plaintiffs assert, works as a trade dress database that can recall and recreate the elements of each artist's trade dress," writes Orrick in a section of the ruling about Midjourney, later stating: "the combination of identified elements and images, when considered with plaintiffs' allegations regarding how the CLIP model works as a trade dress database, and Midjourney's use of plaintiffs' names in its Midjourney Name List and showcase, provide sufficient description and plausibility for plaintiffs' trade dress claim." In other words: the fact that Midjourney used artists name as well as labeled elements of their works to train its model may constitute copyright infringement. But, as I've argued before -- from my perspective as a journalist, not a copyright lawyer nor expert on the subject -- it's already possible and legally permissible for me to commission a human artist to create a new work in the style of a copyrighted artists' work, which would seem to undercut the plaintiff's claims. We'll see how well the AI art generators can defend their training practices and model outputs as the case moves forward. Read the full document embedded below:

Artists' lawsuit against generative AI makers can go forward, judge says | TechCrunch

A class action lawsuit filed by artists who allege that Stability, Runway, and DeviantArt illegally trained their AIs on copyrighted works can move forward, but only in part, the presiding judge decided on Monday. In a mixed ruling, several of the plaintiffs' claims were dismissed while others survived, meaning the suit could end up at trial. That's bad news for the AI makers: even if they win, it's a costly, drawn-out process where a lot of dirty laundry will be put on display. And they aren't the only companies fighting off copyright claims -- not by a long shot.

US Court Upholds Artists' Copyright Claims Against AI Companies

A United States Court dismissed Stability AI and Midjourney's motion to reject the artists' copyright pleas, stating the nature of such cases depends on how the AI tools work along with their implementation by others. The Court has called for evidentiary basis testing of such cases. The US District Court of California on August 12, 2024, approved copyright proceedings against major AI companies Stability AI, Midjourney and DeviantArt after a year-long review of the complaint filed by three independent artists. According to the artists, the companies used their copyrighted works to train their generative AI models. This resulted in the United States Court rejecting the AI companies' motions to dismiss the artists' copyright pleas. In January 2023, three artists Sarah Andersen, Kelly McKernan, and Karla Ortiz filed a class action lawsuit against Stability AI, Deviant Art and Midjourney. These lawsuits are filed by a group of members with a representative from the group. In their lawsuit, the artists alleged that AI tools made by the companies infringed thousands of artists' works. The three AI companies provide AI tools for generating images based on text prompts. The artists said that Stability AI, DeviantArt, and Midjourney trained the Stable Diffusion tool on billions of copyrighted images scraped from the internet. These images were part of the LAION-5B dataset. The AI companies downloaded and used the images without "compensation or consent from the artists." Stable Diffusion is a text-to-image generative AI model released in 2022 by Stability AI. As per the artists, Stable Diffusion made output images by using their artistic works as "training images". The resultant AI-generated images looked like they were "in the style of" the training images. In April 2023, Stability AI sought an order in the Californian Court to dismiss all the claims made by the artists against the company. It said that the artists couldn't prove or produce at least one work of copyright infringement. Stability AI also stated that the artists had not registered their copyrights before filing the lawsuit. Furthermore, the company alleged that the artists failed to identify any particular output substantially similar to any copyrighted work, registered or otherwise. Read more about Stability AI's arguments against the complaint here. The artists added Runway AI to their original complaint alleging that the company "worked with, helped train, and then distributed Stable Diffusion" with Stability AI. They stated that Runway AI made a text-to-image generator available via its online image product AI Magic Tools. The complainants, with additional artists joining the lawsuit, have made the following accusations: District Judge William H. Orrick refused to dismiss the copyright infringement allegations against the four companies and dismissed other allegations related to DMCA violations, breach of contract, and unjust enrichment. The Court denied Midjourney's motion to dismiss the Lanham Act claims. Stability argued that the artists claim that the company is inducing infringement by distributing Stable Diffusion models and when any third-party downloads, uses, or deploys such models. The company stated that the complainants fail to "allege any facts that Stability promoted the use of Stable Diffusion to infringe copyright, as shown by clear expression or other affirmative steps taken to foster infringement". The Court said that the nature of copyright infringement claims (direct or induced) depends on how the Stable Diffusion works and how others implement it. The artists highlighted in their complaint a statement by Stability's CEO that Stability took 100,000 gigabytes of images and compressed it to a two-gigabyte file that can "recreate" any of those images. Stability argued that the statement does not prove intent to foster infringement. The Court decided: "...this is a case where plaintiffs allege that Stable Diffusion is built to a significant extent on copyrighted works and that the way the product operates necessarily invokes copies or protected elements of those works. The plausible inferences at this juncture are that Stable Diffusion by operation by end users creates copyright infringement and was created to facilitate that infringement by design. In addition to the comment of Stability's CEO, plaintiffs reference articles by academics and others that training images can sometimes be reproduced as outputs from AI products. Whether true and whether the result of a glitch (as Stability contends) or by design (plaintiffs' contention) will be tested at a later date. The allegations of induced infringement are sufficient." Runway AI argued that the artists have filed to show acts of infringement by third parties that use the company's products and that the company promoted the use of Stable diffusion to infringe. The Court reiterated the complainants' allegations that Runway "helped train and develop Stable Diffusion" and thus was aware that the product allegedly "uses or invokes the training images in its operation". The Court remarked that these allegations and claims that Runway 'actively induces' others to download Stable Diffusion warrant infringement proceedings. The company distributed the AI tool through coding websites and as part of its AI Magic Tools. The artists claimed that Midjourney trained its product on LAION400M and LAION5B datasets that contain their registered works. They also said that the company incorporated Stable Diffusion into its product. Midjourney said the artists must identify "specific, individual" registered works of each artist that the company used for training. The artists need not present details of their claims due to the datasets' size and the products' nature. The transparency of the open-source software that runs Stable Diffusion is also an issue in this matter, the Court said. The Court explained, "...they are relied on to support the plausibility of plaintiffs' copyright theories (that all or most of the works in the LAION datasets were used by Midjourney and the other defendants to train their AI products, and that plaintiffs' works or their protected elements that are contained in the AI products as the works or protected elements can be recreated by using the AI products). The FAC (first amended complaint) allegations and the exhibits help plaintiffs cross the plausibility threshold". According to the Court, whether plaintiffs can prove their claims is a different matter. The Court advocates an evidentiary basis for testing such claims. DeviantArt argued that the artists have not directly accused it of training an AI model using their works. However, they are objecting to the use of Stability AI and other entities' AI tools. The company argued that holding it liable for using the AI tools would make millions of third parties who used AI products liable for infringement. The artists said anyone can invoke the copies or protected elements of the works using the Stable Diffusion versions. This is because their works remain available in the AI tool in some format or other. Furthermore, DeviantArt said that any use of the artists' works in Stable Diffusion 1.4 is fair use. This is due to the datasets' size and the plaintiffs' inability to generate the copyrighted works using Stable Diffusion 1.4. The Court mandates "evidentiary basis" testing of issues like DeviantArt's DreamUp violating the Copyright Act by reproducing the artists' works. This also includes whether the fair use is applicable in this case. Citing these observations the Court denied the motion to dismiss the Copyright Act claims.