This insightful analysis by Megan Bannigan, Christopher S. Ford, Samuel J. Allaman, and Abigail Liles examines a pivotal February 2025 Delaware District Court ruling in Thomson Reuters v. ROSS Intelligence, where the court granted summary judgment for Thomson Reuters on direct copyright infringement and rejected ROSS’s fair use defense. A key takeaway is that courts may treat unlicensed training of AI models on copyrighted works as non-transformative, with commercial use and market harm outweighing internal-only access arguments. Timing matters: this decision offers a strong early signal that fair use defenses in AI-focused copyright litigation may face steep judicial scrutiny. For IP counsel and AI developers, the article offers actionable clarity on how fair use factors—especially purpose, transformation, and market impact—are being applied in emerging AI cases, making it essential reading for legal professionals navigating generative AI risk.