Content
summary Summary

A group of authors led by Pulitzer Prize winner Michael Chabon has filed suit against Meta and OpenAI in federal court in San Francisco. Another, you might rightfully ask.

Ad

The allegations are the same as in the pending lawsuits: direct and vicarious copyright infringement, removal of copyright information, unfair competition, and negligence.

The authors allege that their copyrighted works have been included in the training material of the respective AI systems without authorization, specifically in the so-called book datasets.

According to the filing, there is no evidence of this, but there is "information and belief". The plaintiffs cite ChatGPT's ability to write detailed summaries of their books as an indication that they were part of the training data.

Ad
Ad

However, these summaries could also be based on summaries from the Internet and would not necessarily mean that the system was trained with the complete books.

OpenAI seeks court clarification on fair use

In a response to a nearly identical lawsuit, OpenAI did not deny (but did not confirm) that the plaintiffs' books were part of the training material. Instead, the company argued that it was making fair use of the data to develop new products. This is allowed under copyright law, even without the authors' consent.

However, OpenAI denied all other allegations, such as that the copyright notices had been removed, presumably with the intention of going to court and obtaining a resolution of the most fundamental question that could make similar AI litigation obsolete: Is it fair use to use copyrighted works to train artificial intelligence?

With these lawsuits against Meta, OpenAI, and Google Deepmind, and similar unanswered questions surrounding the use of code and images for AI training, it is clear that this question needs a fundamental court ruling for everyone to move forward.

To provide peace of mind, Microsoft even offers to cover any legal costs for its customers should a lawsuit arise as a result of working with its generative AI offerings.

Ad
Ad
Join our community
Join the DECODER community on Discord, Reddit or Twitter - we can't wait to meet you.
Recommendation
Support our independent, free-access reporting. Any contribution helps and secures our future. Support now:
Bank transfer
Summary
  • A group of authors, including Pulitzer Prize winner Michael Chabon, has filed suit against Meta and OpenAI for alleged copyright infringement.
  • The authors believe that their copyrighted works have been used in AI training without their consent, although there is no concrete evidence of this.
  • OpenAI argues that the use of copyrighted works in AI training could be considered "fair use" and is seeking a final resolution of the legal issue in court.
Online journalist Matthias is the co-founder and publisher of THE DECODER. He believes that artificial intelligence will fundamentally change the relationship between humans and computers.
Join our community
Join the DECODER community on Discord, Reddit or Twitter - we can't wait to meet you.