liveBook · Manning

Copyright infringement

As explained in LLMs in Production

Unauthorized use, reproduction, or distribution of a copyrighted work.

Copyright infringement becomes a practical risk when an LLM reproduces protected material, imitates a creator’s style, or distributes training-derived content without consent. The concern extends beyond creativity: copied proprietary information can compete with the original publisher and reduce its revenue. section 12.2.1

The New York Times v. OpenAI

The New York Times alleges that OpenAI trained chatbots on its intellectual property without consent and that users received word-for-word responses matching paywalled articles. The dispute illustrates competing concerns: copyright liability may constrain AI development, while unrestricted copying may damage journalism and other rights-holder businesses. section 12.2.1

How LLMs in Production covers it

How deep it goes
Connects unauthorized training, style imitation, plagiarism, and verbatim output to legal and economic consequences without resolving liability.
Examples
Uses alleged New York Times paywalled-article reproductions and creator-style imitation as concrete examples.
What you can do after
Evaluate how training data and generated outputs can expose AI systems, creators, and publishers to copyright risk.
How it connects
Connects copyright infringement with large language models, intellectual property, plagiarism, AI regulation, and journalism.
Cover of LLMs in Production
liveBookmanning.com
Read LLMs in Production on liveBook Get the book at manning.com

Also explained in

Compare how 2 books explain copyright infringement →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of LLMs in Production. The section links point to the parts of the book it draws on.