Looking to interview AI Safety researchers for a book — LessWrong