Neat experiment finds AI fact checks are rated as more helpful & less ideological than human ones "LLM-generated Community Notes can achieve broader cross-ideological acceptance than human-written notes, receiving more positive ratings from raters across the political spectrum"