The legal landscape remains unsettled. No single statute or regulatory body has yet established binding rules for AI training data in fashion, leaving brands to navigate fair use doctrine and emerging industry norms. Copyright liability for generative AI systems remains contested in active litigation, and the scope of permissible training data sources—particularly for copyrighted designs—has not been definitively resolved.
Attorneys advising fashion clients should monitor three areas: first, the outcome of Getty Images v. Stability AI and related cases, which will likely establish baseline copyright liability for AI training; second, whether brands face IP claims for using competitor designs or archived collections to train systems; and third, emerging contractual standards around data licensing, as The Fabricant's ethics framework suggests industry players are moving toward explicit permission requirements ahead of any legal mandate. Brands adopting AI should document their training data sources now, as retroactive compliance may prove costly.