AI as New Epistemic Gatekeeper

If large language models become widely trusted, they will function as de facto fact‑checkers and curators of consensus, replacing some roles now occupied by journalists, tenure systems, and public intellectuals. That shift will reallocate influence — exposing some elites while protecting or entrenching others depending on training data and governance choices. — Changing who the public trusts (algorithms instead of institutions) would reshape political accountability, polarization dynamics, and cultural authority across media, universities, and party politics.