- Adobe accused of using pirated books to train SlimLM
- The company says it used SlimPajama-627B, an open source Cerebras dataset
- The plaintiff claims sufficient financial resources to “vigorously” pursue this case
Adobe will face an AI copyright lawsuit in the US, with a class action case alleging the company trained its AI models on pirated books without permission.
Oregon author Elizabeth Lyon brought the case, alleging that the tech giant had trained its AI models not only on her books, but also on the work of others.
The lawsuit focuses specifically on Adobe’s SlimLM small language models that are used for document assistance tasks on mobile devices.
Adobe Faces AI Training Data Class Action Lawsuit
The company has denied the allegations and claims that SlimLM was trained on SlimPajama-627B, an open source dataset that was released by Cerebras in 2023. However, the lawsuit claims that SlimPajama is a derivative of RedPajama, which allegedly includes Books3, a dataset of nearly 200,000 pirated books.
In summary, Lyon maintains that because SlimPajama includes RedPajama/Books3, it contains copyrighted works without consent, credit or compensation.
Adobe is also accused of having “repeatedly downloaded, copied and processed those works during pre-processing and pre-training of the models.”
This is not the first time that RedPajama or Books3 have been involved in legal cases, having previously appeared in lawsuits against Apple and Salesforce.
Lyon says she is “committed to vigorously pursuing this action on behalf of the other class members” and has the “financial resources to do so.”
The plaintiff seeks “statutory and other damages,” reimbursement of attorneys’ fees, and a declaration of willful infringement from Adobe.
TechRadar Pro has sought a formal response from Adobe, but the company has not yet responded.
Follow TechRadar on Google News and add us as a preferred source to receive news, reviews and opinions from our experts in your feeds. Be sure to click the Follow button!
And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form and receive regular updates from us on WhatsApp also.
