Committee considers ‘ingredients label’ for AI training data; supporters and tech groups urge clarity on trade secrets
HB 2,503 would require developers to post high-level documentation of data used to train generative AI systems; sponsors called it an 'ingredients label' for models, advocates cited safety concerns about problematic training data, and industry pressed protections for trade secrets and narrow applicability.
The full story
The committee heard House Bill 2,503 on Jan. 21, which would require developers publishing generative AI systems to post high-level documentation about training datasets — sources, approximate data point counts, whether copyrighted material was included, and whether datasets were licensed or purchased — before making systems publicly available in Washington.
Sign up to keep reading.
Your free newsletter brings you what your city council, county, and school board discussed and decided.