ChristianKl comments on Restricting Dangerous Research: Has It Worked Before, and Could It Work for AI?