Stability AI allegedly used copyrighted artist works to train Stable Diffusion
Stability AI faced multiple lawsuits alleging the unauthorized use of billions of copyrighted images for training Stable Diffusion. These legal challenges centered on the use of datasets like LAION-5B which scraped content from the internet without artist consent.
The training process relied on scraping billions of images without a legal framework for artist compensation or consent.
Key facts
- What
- Stability AI faced multiple lawsuits alleging the unauthorized use of billions of copyrighted images for training Stable Diffusion.
- Incident date
- Dec 1, 2022
- Who
- Stability AI
- Failure mode
- Policy Violation
- AI surface
- Media Generation
- Severity
- Medium
What happened
A class action lawsuit and other legal challenges alleged that Stability AI used copyrighted images from artists without permission to train Stable Diffusion. These claims focused on the use of the LAION-5B and LAION-400M datasets. While the UK High Court rejected some secondary copyright claims in the Getty Images case in 2025, the allegations regarding training data usage sparked widespread legal disputes.
What broke inside the model
- 01 · TriggerA prompt pushes against a deployment boundary.
- 02 · Model stepThe model produces the disallowed output.
- 03 · Control gapNo enforcement blocks it at generation time.
- 04 · FailureThe output crosses the policy line.
- 05 · ConsequenceA limit the business set is breached in public.
The output crosses a policy boundary the deployment had defined.
The failure occurred in the data procurement process where copyrighted material was ingested into training sets without a licensing framework. The model training mechanism relied on these scraped datasets, creating a conflict between AI development and intellectual property rights.
What it cost
Sources
- Court FilingAndersen v. Stability AIbakerlaw.com
- Court FilingGetty Images v Stability AIjudiciary.uk
Cite this entry
https://failureindex.ai/failures/stability-allegedly-used-copyrighted-artist-worksAI Failure Index. "Stability AI allegedly used copyrighted artist works to train Stable Diffusion" (FI-0411). Realm Labs. https://failureindex.ai/failures/stability-allegedly-used-copyrighted-artist-works (indexed Jun 10, 2026).Data fields CC-BY 4.0, prose citation permitted. Incident ID FI-0411. Full dataset at /data.
Note from Realm Labs, the Index steward
How Realm would have caught this
- Prism
- OmniGuard
Realm compares what the model is about to output or do against the policy that governs the deployment, in real time, and can deny or redact the action before it takes effect, which is the gap an after-the-fact review never closes in time.