To many users, AI seems like magic. It can generate images, write papers, and summarize complex topics into tidy bullet points. But what sources are under AI’s hood? How does it know so much? AI models are trained on vast amounts of human-generated content from the internet – including everything from transcribed YouTube videos and Reddit threads to image libraries and the entire text of Wikipedia. This raises a critical legal and ethical question: Are AI companies committing copyright infringement by scraping and using this content without permission? The New York Times, numerous authors, artists, and musicians think so. They’ve filed lawsuits against companies like OpenAI and Meta, arguing that their work fuels these models – yet they receive no credit, attribution, or compensation. AI companies defend their practices by citing the Fair Use doctrine, claiming that AI outputs are “transformative” – computational mashups that are fundamentally different from the original works. But many legal questions remain unresolved, and frustration is building across the many creative industries that are impacted. As students increasingly turn to AI, it’s important for them to consider not just what the tools can do, but the human contributions and intellectual property that made them possible. Understanding the legal and ethical issues is key to becoming a responsible and informed user.