Raw LLM Responses

Y'all - someone is really going to have to blackpill me on this (is that the right word?). Because LLMs - even Grok - afaict universally provide information that is broadly accurate and inline with institutional sources and guidance. We've had a major crisis in information literacy and trust of institutions. I can't help but think of ChatGPT as a backdoor way to get people advice and information that is broadly aligned with expert recommendations and the evidence when so many people have upside down information literacy - that is, that the more reputable a source, the less they believe it, and the less reputable a source, the more they believe it. Is going to ChatGPT better than going to your local domain expert? Usually not, though.. like sometimes lol. Is it better than going to your anti-establishment influencer who's got a bajillion followers? Definitively yes. So.. somebody connect these dots for me. How is it not so that a low info literacy rando with ChatGPT is MUCH better than a low information rando with tiktok, podcasts, and a search engine.