Yes, you can sue an AI company for using your data without consent in some situations. Success depends on what data was taken, how it was obtained, and what law protects it.
AI training means feeding text, images, or other material to a model so it learns patterns. Copyright owners have one path to court. People whose personal information was misused have another.
Table of Contents
- When have copyright owners won?
- Why does the source of training data matter?
- What if the company used your photos or personal information?
- What can you do now to limit AI use of your data?
When have copyright owners won?
Copyright owners can sue when a company trains on their protected works without permission. The Third Circuit affirmed that ROSS infringed 2,243 Westlaw headnotes and rejected competitive AI training as fair use, according to LawNext in its summary of the unsealed opinion.
That case involved headnotes, which are short legal summaries. The court treated the dispute as ordinary copyright infringement. It did not create a special exemption for AI.
Why does the source of training data matter?
How the company got the material can decide the case. Training on lawfully obtained books can be transformative fair use, but copying pirated copies is different.
A federal judge approved a $1.5 billion class settlement with Anthropic covering about 482,000 books, or roughly $3,000 per work before fees, as reported by the Associated Press via BusinessMirror in its report on the approved settlement. The payment followed claims tied to pirated library copies used for training.
What if the company used your photos or personal information?
You may have a deception or privacy claim if a company broke its privacy promises. The Federal Trade Commission alleged that OkCupid shared nearly 3 million user photos plus demographics and location data with Clarifai in 2014, as described in the FTC case record.
That type of claim focuses on broken promises, not copyright. It applies when a site says it will protect personal information and then shares it for AI development.
What can you do now to limit AI use of your data?
Start by using rights you already have. Californians can invoke rights to know, delete, and opt out of the sale or sharing of personal information, then file business requests to limit future AI-data use, according to the California Attorney General's privacy guidance. You can also change product settings where available:.
- Turn on "do not train on my content" or Data Controls for applicable chat tools
- Check whether new conversations are excluded after you opt out
- Use a business or API tier if you need inputs excluded by default
- Keep copies of opt-out confirmations and deletion requests