An American Answer to China's Open AI Models: A Safe Harbor for Open Weights

manhattan.institute