On July 9, 2026, the World Intellectual Property Organization (WIPO), the World Health Organization (WHO) and the International Telecommunication Union (ITU) released AI-enabled Health Innovation and Intellectual Property: From Idea to Impact. The guide brings IP strategy, commercialization, health-data governance and technical standards into one framework for AI-enabled healthcare projects. It treats training-data provenance, patient consent, cross-border transfers, open-source licensing, model weights and technical documentation as connected business risks, while stressing that patents, trade secrets, copyright and contracts often need to work together.
For patent planning, the guide encourages applicants to describe concrete technical implementations in training pipelines, privacy-preserving methods and hardware-software interaction, and to use the PCT route when scaling into priority markets. European filings still need to reflect EPO practice on the “technical character” of AI and machine-learning inventions. The document does not create a single global legality test for using medical datasets or copyrighted literature in model training. Its practical message is narrower and more useful: data permissions, patent filings and regulatory access should be designed in parallel. A startup that leaves data provenance until due diligence may discover that a strong algorithm patent does little to rescue a product whose licensing chain cannot support commercialization in Europe.
