generative AI truly contributes to productivity. While it can serve as a tool to speed up repetitive tasks or the early stages of idea exploration, doing so often requires additional time and resources for verifying the accuracy of AI-generated outputs, correcting biases or errors, and developing the skills needed to use these tools effectively. Efforts are also needed to reorganize work structures so that the adoption of technology does not undermine the development of workers’ skills and capacities. In this sense, generative AI should be understood as part of an organization’s broader digital transformation process. As illustrated by a survey finding that 95% of companies investing in generative AI have not achieved net organizational gains from it, this transition is far from straightforward.1 Copyright Issues in Training Data and Creative Labor Developing generative AI models requires access to vast amounts of data. This includes not only texts, images, and code published on the web, but also copyrighted works such as books and other published materials. In many cases, AI companies have neither sought explicit consent from rights holders nor provided compensation for the use of such works. These practices have fueled tensions between industry claims of “fair use” and concerns over the infringement of creators’ rights. At 1 A  ditya Challapally, Chris Pease, Ramesh Raskar, Pradyumna Chari. The GenAI Divide: State of AI in Business 2025. MIT NANDA. Generative AI Guide for Civil Society