Rather, I think they're afraid of the criticism they'll face over such biases if they make the model generally available. It doesn't take much imagination to s

Rather, I think they're afraid of the criticism they'll face over such biases if they make the model generally available. It doesn't take much imagination to see these issues being used by AI opponents as justification to ban or heavily regulate machine learning research.