Fair use
Fair use allows limited use of copyrighted material without permission for purposes like criticism or education.
Fair use is a legal idea in the United States that lets people use parts of copyrighted material, like text or images, without asking the owner first. This is allowed only for specific reasons, such as commenting on the work, reporting the news, teaching, or doing research.
For example, if a news report shows a short clip of a movie to discuss its themes, that could be considered fair use. It's not a free pass to use whatever you want; the use must be limited and serve a purpose that benefits society, like education or commentary.
Fair use is a defense against copyright infringement claims under U.S. law. It permits the unlicensed use of copyrighted material for purposes such as criticism, comment, news reporting, teaching, scholarship, or research. Courts analyze four factors to determine if a use is fair: the purpose and character of the use, the nature of the copyrighted work, the amount and substantiality of the portion used, and the effect of the use upon the potential market for or value of the copyrighted work.
In the context of AI, fair use is central to debates about whether training data scraped from the internet, which often includes copyrighted material, infringes on existing copyrights. Companies argue that training AI models constitutes fair use, while creators and publishers often contend it does not.
Fair use is a key legal concept in ongoing debates about AI training data. Lawsuits, like the one brought by The New York Times against OpenAI, directly question whether AI companies' use of copyrighted material for training constitutes fair use.
- OpenAI and Microsoft knew they were starting a ‘doom loop’ for the webOpenAI · The Verge AI ·
- OpenAI and journalismOpenAI · OpenAI News ·
- OpenAI ChatGPT
- OpenAI has defended its AI training practices as fair use in response to lawsuits, while also exploring partnerships with publishers.
- Public Domain
- Material not protected by copyright and can be used freely by any
- Creative Commons
- A system of licenses that allow creators to share their work u
- Infringement
- Using copyrighted material without permission or legal justificati
The concept of fair use has roots in English copyright law dating back to the 18th century and was codified in the U.S. Copyright Act of 1976. Source ↗
Find a news article or blog post that discusses a recent movie or book. Then, try to find a review or commentary that quotes short passages from the work being discussed. Consider how the quotes are used to support the reviewer's points and whether they are essential to understanding the commentary.
The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.
- PromptRole-play a student struggling with math for AI tutor simulationSimulate a student's math struggles and uncertainty for AI tutor evaluation. · 5 min
- TemplateReusable workflow templates for healthcare teamsTeams can create and share templates for common workflows like documentation or prior authorizations. · 15 min
- TemplateTemplate for building an AI tool tailored to your workA structured approach to defining and building a custom AI tool for specific business needs. · 15 min
- TemplateEvals CSV prompt list for agent skillsA CSV file template for listing prompts used in evaluating agent skills. · 20 min
- HeuristicDouble coding agent task completion with docs linksProviding coding agents with a documentation link significantly increases their task completion rate. · 2 min
Questions people ask
- Is all use of copyrighted material fair use?
- No, fair use is a limited exception. It depends on the specific purpose, nature of the work, amount used, and market impact.
- Does fair use apply to AI training data?
- This is a major legal debate. AI companies argue it is fair use, while many copyright holders disagree and are pursuing legal action.
First seen on Subvolts Jan 8, 2024. Updated Sep 25, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.