What the Luddites Can Teach Us About Societal Response to AI — LessWrong