LLMs: and the death of critical data gathering?

Photo by Mick Haupt on Unsplash

It’s no secret that LLMs have made some of the repetitive and boring tasks a million times easier - no mood to write an essay? No problem, just ask ChatGPT BingAI or any of the 100 other flavors.
With Bing AI and now with the recent power of chatGPT to even surf the web, it only made it easier to even search and write about the latest info and real-time data, and not be limited by some obscure training data bias. But what does this all mean to an average person, probably nothing, but at least personally on a tech enthusiast level, this, is a kinda (platform shattering experience) because this was a necessary skill to survive, the ability to find very cliched or not so easily available information over the internet, this would sometimes take meticulous googling with specific keywords, limiting searches to certain domains, etc.
To give an example, I was having an issue with a particular WiFi driver on Linux which issue was so specific that it would take me sometimes hours to find a suitable alternate driver, and it was a repetitive task, every kernel upgrade would break it, and I had to search it up again, rebuild the driver, etc.
After a few months of this, I had refined the solution to some trusted devs and their repos, and although my issue was specific to my device model, other models had similar issues, and very often the blogs, etc wouldn’t have mentioned the model or not specific enough and mention the range, which would cause further breaks and re-trials. But somehow, Bing AI resolves this issue by always getting the correct hit in the top 3, I guess that is the power of web indexing.
So is this the end, or a stepping stone to broaden the horizon of data gathering to extents previously unknown? Is this the next step in the expansion of info hunting, like over the past few decades, has shifted from meticulously searching books or other printed materials to almost completely digitally?