

Strengthening the Public Data Layer for AI
The workshop will bring together leading AI researchers, model developers, and data infrastructure experts to help shape a technical roadmap for dismantling the barriers that prevent academic researchers, startups, and government agencies from working with large-scale open training data and extending the linguistic representation of these datasets.
Core focus areas will include:
Reducing barriers for new entrants to train large language models on open-source datasets through cross-cloud storage and hosting, enhanced crawling with modularity and adaptive management, and improved dataset usability via tooling and metadata.
Extending the public AI training data language corpus by strengthening Common Crawl’s multilingual classification and growing the ecosystem of foreign language datasets.
Identifying emerging challenges and opportunities to extend U.S. training data leadership for open source AI.
Convened by Renaissance Philanthropy in collaboration with the Common Crawl Foundation, with support from the U.S. National Science Foundation.