“Clem Delong argues that governments should regulate frontier AI labs and open-source AI differently, as open-source models are more transparent, specialized, and pose fewer risks.”