AI is going to cause a regression to the most anodyne output across many industries. As humans who had to develop analytical skills, writing skills, etc., we struggle to imagine the undeveloped brains of those who come of age in the zero-intellectual-gravity world of AI. OpenAI's study mode is at best a fig leaf.
edit: this comment was posted tongue-in-cheek after my comment reflecting my actual opinion was downvoted with no rebuttals:
I would say the modern digital world itself has already had the bigger impact on human thinking, at least at work.
It seems with computers we often think and reason far less than without. Everything required thought previously, now we can just copy and paste out word docs for everything. PowerPoints are how key decisions are communicated in most professional settings.
Before modern computers and especially the internet we also had more time for deep thinking and reasoning. The sublimity of deep thought in older books amazes me and it feels like modern authors are just slightly less deep on average.
So then LLMs are in my view an incremental change rather than a stepwise change with respect to its effects on human cognitive.
In some ways LLMs allow us to return a bit to more humanistic deep thinking. Instead of spending hours looking up minutia on Google, StackOverflow, etc now we can ask our favorite LLM instead. It gives us responses with far less noise.
Unlike with textbooks we can have dialogues and have it take different perspectives. Whereas textbooks only gave you that authors perspective.
Of course, it’s up to individuals to use it well and as a tool to sharpen thinking rather than replace it.