When LLM judges agree, should we believe them?
An article on Hacker News examines whether consensus among large language models can be trusted as a form of judgment. It discusses the mechanisms behind model agreement, potential biases, and the limits of relying on AI for decision‑making. The piece highlights the need for human oversight and rigorous evaluation before accepting model consensus as authoritative.