Concern 707 · Working draft · editorial review in progress
Training-data compensation for artists
Training-data compensation concerns whether and how artists are paid when their work is used to develop AI systems that may compete with their own work.
Understanding the concern
A model may learn patterns from large collections of images or other creative material. Artists may not know whether their work was included, who authorized the use or how any payment is calculated. The economic question is separate from whether a particular generated image closely copies a particular original.
Copyright and licensing debates address several different uses, and legal outcomes depend on the facts and jurisdiction. A licensing scheme also raises practical questions about transparency and who receives money. Think of a library built from many contributors: paying the library's owner does not automatically show that each contributor was fairly represented or rewarded.
An example
An artist learns that a collection containing their work was licensed for training but cannot determine whether the agreement covers them or pays them.
Illustrative unless identified as a documented case.
Words in plain English
- AI
- Software that learns patterns from examples or uses rules to perform tasks such as recognizing images or generating text. It can make mistakes and does not understand every situation as a person would.
- Model
- The learned part of an AI system. It uses patterns found during training to produce an answer or prediction.
- Training
- The process of using examples to shape how an AI model behaves. Learning patterns from material does not establish that the material is true or fair.
Evidence and its limits
The Copyright Office report analyzes law and policy and is labelled pre-publication. It does not impose a universal payment rule or resolve every training dispute.
US Copyright Office · May 2025 pre-publication version · Legal policy report
Copyright and Artificial Intelligence, Part 3: Generative AI Training ↗ (opens in a new tab)Training, fair use and licensing analysis. Not a court judgment, universal licensing mandate or final resolution of every pending case.
Broader frameworks and background
These resources provide context. They do not establish every claim in this concern.
A question to discuss
What would make consent, accounting and payment understandable and fair to the individual creators involved?
Keep track of different views, what needs checking, and any next step in your own discussion notes.
Open discussion notes →Principles for the conversation
Before you finish · reflect together
Virtues and principles to consider
Virtues are qualities we practice in how we treat people and make choices. They are not scores for judging someone. Use these ideas to explore what a thoughtful response could look like.
Seeing potential
Seeing potential means recognizing people's capacity to learn, contribute, and grow. It resists reducing someone to a score or to the tasks a machine can perform faster.
Use tools to support people's development. Give learners time, feedback, and real opportunities to practice. A prediction about someone's performance should not close off their future.
Ask together: How could this help people develop abilities they value?
Solidarity
Solidarity means standing with people whose lives are affected, especially when they have less power to protect their interests. It turns concern into a willingness to share responsibility.
Include affected people in decisions and support practical remedies when something goes wrong. Do not leave the cost of learning or repair entirely with those who were harmed.
Ask together: What would it mean to stand with the people carrying the greatest burden?
Dignity
Dignity means recognizing each person's worth regardless of usefulness, status, or a system's score. People deserve respect and a voice in matters that affect them.
Explain important decisions, avoid humiliating treatment, and keep room for a person's story. Do not treat a prediction as a complete account of who someone is.
Ask together: Does this use treat each person as someone whose experience matters?
Editorial ratings
These are editorial judgments from the register, not measured probabilities or predictions about your situation.
About the register and scoring →More ways to explore
Search for more writing
For more specific or recent material, open a search below or ask for a web summary. AI-generated summaries can contain mistakes; open the original articles to check them.