If you feed a bunch of a target's personal info into an LLM, would it guess their password more quickly than a human? What about an LLM trained specifically for

If you feed a bunch of a target's personal info into an LLM, would it guess their password more quickly than a human? What about an LLM trained specifically for the task? It could be the end of non-random passwords.