10 months ago
George R.R. Martin's Copyright Lawsuit Against OpenAI Gains Momentum
Imagine if someone used your drawings without asking to teach a robot how to draw like you.
That's kind of what famous authors like George R.R. Martin are saying about companies that make AI.
They say these AI companies used their books, which are like their special drawings, to teach their computer programs, like ChatGPT, how to write stories.
The computer then wrote a story that sounded a lot like the author's own work.
A judge looked at this and said it's possible a jury could agree that this is against the rules, like breaking copyright.
So, the case can move forward to see if the AI company did anything wrong by using the authors' books to learn.
A federal judge allowed class-action lawsuits against OpenAI and Microsoft to proceed.
The lawsuits allege that AI models were trained on copyrighted works without permission.
George R.R. Martin's lawyers provided an example of ChatGPT generating text similar to his 'A Song of Ice and Fire' series.
U.S. District Judge Sidney Stein stated a jury could find the AI's outputs substantially similar to plaintiffs' works.
Prominent authors including Michael Chabon and Sarah Silverman are part of the lawsuits.
- Who
- George R.R. Martin, Michael Chabon, Ta-Nehisi Coates, Jia Tolentino, Sarah Silverman, OpenAI, Microsoft, U.S. District Judge Sidney Stein
- What
- A federal judge allowed class-action lawsuits alleging copyright infringement against OpenAI and Microsoft to proceed, citing that AI-generated text imitating 'A Song of Ice and Fire' could violate George R.R. Martin's copyright.
- Where
- Manhattan federal court
- When
- Monday (referring to the judge's ruling)
- Why
- Authors allege their copyrighted works were used without permission to train large language models, leading to AI-generated content that imitates their original creations.
Authors' Allegations
OpenAI/Microsoft Defense (Implicit)
Use of Copyrighted Material
Authors' Allegations
Authors claim their copyrighted works were used without permission to train AI models.
OpenAI/Microsoft Defense (Implicit)
The defense may argue that training AI on publicly available data falls under 'fair use', a legal exception to copyright.
AI Output Similarity
Authors' Allegations
AI-generated content is substantially similar to the authors' original works, infringing on their copyright.
OpenAI/Microsoft Defense (Implicit)
The AI is a tool that produces new content based on patterns learned, not a direct copy; similarity does not automatically equate to infringement, especially if 'fair use' applies.
Key facts
- Judge
- U.S. District Judge Sidney Stein
- Allegation
- Copyright infringement of literary works used to train AI models
- AI Tool
- ChatGPT
- Plaintiff Authors
- George R.R. Martin, Michael Chabon, Ta-Nehisi Coates, Jia Tolentino, Sarah Silverman
- Defendant
- OpenAI and Microsoft
Quotes
U.S. District Judge Sidney Stein
The federal judge presiding over the class-action lawsuits against OpenAI.
“A reasonable jury could find that the allegedly infringing outputs are substantially similar to plaintiffs’ works”
timesnownews.com
