371
Productivity & Workflow357
Software Development253
Automation & Workflow230
Marketing & Growth213
AI Infrastructure & MLOps197
Photography & Imaging170
Writing & Content Creation197
Design & Creative176
Data & Analytics162
Customer Support161
Education & Learning150
Sales & Outreach132
Voice & Speech131
Operations & Admin118
Courts and lawmakers in the US, EU, and Australia are debating whether AI companies can train on copyrighted work without permission, and what creators should get.
In short: Governments and courts around the world are still split on whether AI companies can use copyrighted work to train AI without permission, and what creators are owed.
The biggest copyright fight around AI in 2026 is about training data. Training data is the large pile of text, images, and audio that an AI system studies to learn patterns (like how a student learns by reading many books). The question is whether AI developers can use copyrighted material for this, without asking first.
In the United States, the Copyright Office says current law can already handle some issues. Human-made parts of a work can be protected by copyright, but material created entirely by AI cannot. In a May 2025 report, the office also said “fair use” has to be decided case by case, rather than giving AI training a blanket pass.
US courts are sending mixed signals. One appeals court rejected a fair use argument in a case involving ROSS Intelligence and Westlaw legal content. Other cases have leaned the other way, based on their specific records. These decisions do not settle whether AI training is legal in general.
Lawmakers have not settled it either. In Congress, ideas include requiring AI companies to disclose what they trained on, or adding new penalties or remedies, but recent proposals have not advanced far.
Outside the US, the European debate is heating up. A June 2026 proposal from the Max Planck Institute recommends new AI-related rights and pay systems, and Germany and ten other EU countries have asked for stronger EU rules. In Australia, creative groups are pushing back against changes they worry would let AI companies use works unless creators opt out.
Expect more “tweaks” proposals, like clearer disclosure, simpler licensing, or collective payment systems (similar to how radio royalties are handled). For now, the rules will keep varying by country and by court case.
Source: NYTimes