[SEQ RERUN] Trust in Math — LessWrong