Wikidata Knowledge Graph to Enable Equitable and Validated Generative AI

Explore how open-access knowledge graphs like Wikidata can enhance generative AI applications in this 31-minute conference talk by Jonathan Fraine and Lydia Pintscher from Wikimedia Deutschland. Discover the potential of Wikidata's crowd-sourced, structured data items in over 300 languages to provide authenticated and equitable world knowledge for LLM and ML applications. Learn about practical integration techniques for leveraging Wikidata's open-source knowledge graph to augment and validate generative AI inference. Gain insights into semantic search capabilities and the use of vectorized Wikidata search API to enhance AI applications with diverse, crowd-sourced data, offering a more representative spectrum of human knowledge compared to standard internet corpora.