Two neuroscientists from SUNY Downstate Health Sciences University have recently launched a class-action lawsuit against Apple. The case quickly drew widespread attention as it focused on Apple’s use of copyrighted books in developing its Apple Intelligence artificial intelligence model.
Allegations named Apple’s reliance on data from so-called shadow libraries, sparking renewed debate about the line between innovation and intellectual property rights in artificial intelligence.
What prompted neuroscientists to sue Apple?
Professors Susana Martinez-Conde and Stephen Macknik claimed that Apple utilized works, including their own, without consent to train its AI systems.
The lawsuit cited specific titles, such as "Champions of Illusion" and "Sleights of Mind," as being among the materials allegedly scraped from “shadow libraries.”
The neuroscientists argued that Apple benefited from thousands of copyrighted books, citing substantial commercial gains, and sought monetary damages and an injunction to halt usage.
This action highlighted the growing tension between content creators and technology companies developing AI models.
Did you know?
The Books3 dataset, central to this lawsuit, reportedly hosted over 190,000 books, some in plain text format, and has faced global scrutiny for its role in large-scale AI training.
How did shadow libraries become central to the case?
The case referenced the Books3 dataset, which has faced broad scrutiny for its inclusion of over 190,000 copyrighted works sourced from shadow libraries.
Evidence reportedly suggests that Apple's OpenELM model, utilizing Books3, raises questions about the vetting and legitimacy of AI training data practices.
Shadow libraries, often comprising pirated books, have emerged as a controversial source for machine learning researchers and AI developers.
Legal and ethical issues arise as more lawsuits highlight the widespread and deeply integrated nature of these sources in the evolution of commercial AI.
What is the scope and impact of the lawsuit?
The scope quickly broadened, with the class action potentially covering thousands of authors. The complaint asserted that Apple’s financial gains after its AI launch were directly related to unauthorized use of protected texts, referencing a market value increase of over $200 billion in a single day following the debut of Apple Intelligence.
Legal observers noted this suit as part of a wave of copyright litigation, as writers, scientists, and publishers demand accountability and compensation.
A court ruling or settlement could set new boundaries for data usage in commercial AI development.
ALSO READ | Meta mandates AI integration for 80 percent of metaverse workers
How have previous AI copyright disputes unfolded?
Earlier in September 2025, Apple faced another lawsuit from a group of authors dealing with similar allegations.
Across the industry, tech companies such as OpenAI, Microsoft, and Meta have been drawn into legal battles over copyright issues related to AI training.
The stakes escalated when Anthropic agreed to a historic $1.5 billion settlement with authors, reshaping expectations for AI copyright enforcement.
The most recent decisions suggest that courts are still navigating the application of copyright to large language models, with each significant settlement influencing future legal tactics and defense approaches by technology companies.
Where does this case leave Apple and the tech industry?
Apple responded by reaffirming its commitment to responsible AI development and defending its practices.
However, the public and legal scrutiny is unlikely to diminish as more content creators review what data is used to train commercial AI products and how those sources are obtained.
For the broader technology sector, this high-profile case may become a touchstone in the ongoing debate about responsible data sourcing.
Companies will likely reconsider how they acquire training data, with more transparency and compensation frameworks on the horizon.
Comments (0)
Please sign in to leave a comment