I am a Gen Z university student. The ChatGPT got boom when I was in 12th grade and I have been using ChatGPT, Gemini and Deepseek since then. I am starting to notice that now I cannot do searching anymore. I literally cannot do anything on my own like coding, learning something, fact checking, doing my assignments. I even become so reliant on it that I sometime even ask it what food should I eat, what exercises I should do. Sometimes, I “talk” with ChatGPT more than real human beings, maybe cuz I somehow end up having a bad experience or not so fulfilling experience with people. I have to use AI again and again for the same question, e.g., I asked a question today, I got the answer. If I encountered that question later again, I would not know the answer and have to go to AI for solution. What to do now? How to get out of this dependence on AI?

  • pulsewidth@lemmy.world
    link
    fedilink
    arrow-up
    14
    ·
    22 days ago

    That’s a great question. I tease, but first step is realising the problem and you’ve done that. Much harder is acting on it.

    The truth is nobody really knows. It’s new. Addiction to it is also new as is the research showing its deleterious effects on the brain.

    If I was looking for somewhere to start by trying to read a book a every week or two. Alternate between appealing fiction and an important topic. If you have readings set by your educational institution - do those instead. Reading and digesting plaintext information works a lot of your brain and doesn’t come with any immediate answers to questions, it will help you build your patience back.

    • Naive@lemmy.worldOP
      link
      fedilink
      arrow-up
      4
      ·
      22 days ago

      Trying to do that and yeah, I want to read as much as I can before these techbros end up destroying all the books that we have right now.

      • Final Remix@lemmy.world
        link
        fedilink
        arrow-up
        3
        ·
        22 days ago

        Also, libraries. Local and school. Your librarians are information brokers. Their job is to find the shit you want and may not even be sure exists, but they’ll fuckin’ find it.

        Even short novellas or whatever (Murderbot is crazy popular and they’re like 120 pages); length don’t count for shit, just read. Just get the momentum going, find a genre and/or author you like, read and keep at it.

  • lemmy_outta_here@lemmy.world
    link
    fedilink
    arrow-up
    6
    ·
    22 days ago

    Your brain is very plastic. If you can endure the discomfort of quitting AI cold turkey, you will start to lift out of your fog quickly. I went through a period of severe depression and heavy drug use that lasted so, so long. By the end, I could barely read 5 pages. Things improved and I started reading/writing again. My skills came back, and it didn’t take years. If you are like me, you can bounce back before 2026 is over. Read easy stuff, keep a journal, do simple coding projects - but use no AI. Your momentum will build fast.

  • TootSweet@lemmy.worldM
    link
    fedilink
    English
    arrow-up
    4
    ·
    edit-2
    22 days ago

    Ignore everyone in here who implies there’s even a tiny extent to which this is your fault. It’s 100% not your fault. AI is far from the only difficulty your generation has had to deal with in your education. Things like Tiktok/shorts are also damaging. COVID set a ton of students back quite a lot. The school system is fucked in so many ways. And your devices all try to keep you from doing anything “advanced” with them in favor of prepackaged consumer experiences. It’s extremely unfair to expect students these days to just grin and bear it and somehow manage to retain their brain function and simultaneously eek out some measure of success in this world. And the trend has been toward the worse.

    And yet, your only real option is to do the hard work to reclaim your own brain. But there’s good news. First: you’re in college. Congrats on that! That’s a huge accomplishment in itself! Second: your attention span isn’t so broken as to keep you from reading this far. Third: recognizing that there’s a problem and asking for help and advice is a hugely good sign that you’ve got enough level-headedness to improve your situation.

    I think a lot of people have already said “stop using generative AI.” And I think that’s very much on track. I very much think you’re right in thinking that AI is probably the biggest source of the problems you’re describing. And most of the rest of what people are saying here is good advice. “Excercise it like a muscle.” “Engage with communities like this one.” Etc.

    Honestly, a fair amount of what keeps me avoiding AI is resentment. I don’t know if that’ll work for you, but keep in mind what they did to you on purpose in service to their bottom line. It’s a lot easier to stay away from something if you don’t fall for the false sycophancy and you keep in mind the contempt the companies behind it have for you.

    Find a hobby that you enjoy that takes a lot of “brains”. I write software a ton, but it definitely doesn’t have to be that. Yarncraft is surprisingly thinking-intensive. Games like sudoku. Puzzles. Woodworking. Bonus points if it’s something that gives you something to “show for it” when you’re done.

    I’d go a little further, though. It might be good to also eschew as many addictive dopamine sort of things as possible. Tiktok, addictive games, etc. (There’s a lot of propaganda from #nofap weirdos out there claiming that porn is the worst thing you can do to your brain. I don’t think it’s that simple and I do think probably you’re the best one to make the call for yourself whether that’s the case for you. Maybe “detoxing” from porn for, say, a week at a time could be helpful from time to time, though. Also, I believe strongly that there’s nothing wrong specifically with masturbation or (safe) sex itself. (In fact, it’s probably better for you to do so than not to do so.) If there’s any truth to the #nofap stuff, I think it only applies to porn. And even then, I think that only applies to a minority.)

    Consume more long-term educational content. Read books. Watch stuff that teaches you stuff. (I’m really into STEM stuff. Math, physics, engineering, hardware hacking, programming, etc. I also enjoy a lot of social commentary stuff, philosophy, and psychology. But do what works for you.) Opencourseware is a great resource.

    Further still, I think meditation is a pretty big “cheat code”. It sucks. A lot. For a while. But it gets easier and I think it’ll improve your brain quickly in ways you can’t necessarily even anticipate now. Start with short sessions. 10 minutes, 5 minutes, or even 2 minutes. Here is a decent guide for beginners. Over time, you’ll start enjoying it.

    Good luck! You’re doing good recognizing the problem and planning to take action to remediate it! And you’re more than welcome here. Post updates if you want!

    • Naive@lemmy.worldOP
      link
      fedilink
      arrow-up
      2
      ·
      21 days ago

      I don’t like STEM much even though I am in this field. I also found LibreTexts to be good resource. They have the mission to promote open education

  • whalebiologist@lemmy.world
    link
    fedilink
    arrow-up
    3
    ·
    22 days ago

    I’m going to give you an example, if you don’t understand. Thats ok. It’s normal not to understand things on the first parse:

    "I just saw the odyssey, the main character is great. I don’t remember the actor’s name. He’s in batman or something. I don’t know. " At this point in the conversation everyone will either say “Matt Damon?” or take out their phones, then say “Matt Damon, but he wasn’t in batman, that’s Ben Affleck” “But yeah, he was great and I like the sounds the ships make when they are at sea, very authentic sounding. I had a good time seeing it.”

    Here I will say, knowing the actor’s name has no bearing on the conversation at hand. So instead of looking something up we can just move forward with an irrelevant gap. This is what will cure you of your sickness. You are uncomfortable not knowing something, maybe you worry people will think you are stupid. In truth, it is smarter to admit you don’t have a complete understanding of the situation. Everyone in my example who took out their phones is also making a grave error. They equate their ability to reference IMDB’s cast list with “knowing” something. Stop looking everything up, reading does not equate to knowing - even when you are getting things from a reliable source.

  • Buddahriffic@lemmy.world
    link
    fedilink
    arrow-up
    3
    ·
    22 days ago

    For coding, try going back to square 1 or starting a new language from scratch. Like finding tutorials that start with building a “hello world” app and building it up from there. If you already had some coding skills before using AI, it should start coming back to you or at least be familiar.

  • red_tomato@lemmy.world
    link
    fedilink
    arrow-up
    2
    ·
    22 days ago

    AI is like a slot machine. You pull the lever. You see it do the magic. You get the dopamine boost. You do it again.

    It can be addicting.

    I use AI quite a lot at work for that productivity boost. I never use AI outside work though. What works for me is that I’m quite used to interacting with my own inner monologue. I rarely go out and seek answers to my questions. Those times I do seek answers I accept I won’t find the answer. I’ve made peace with not having the answer to everything.

    Practice being alone with your mind. Go for long walks. Learn to accept it’s ok not to know.

  • OrteilGenou@lemmy.world
    link
    fedilink
    arrow-up
    2
    ·
    22 days ago

    Instead of asking, try to figure out what you think it will say first

    If you’re right, look up the opposite.

    If you’re wrong, find out why, and try to figure out what the reasoning is.

    Eventually, don’t look it up. If you hear something and you know you’re right, trust it. If you have the slightest doubt, look up the opposite first. If the opposite of what you thought it was comes up, you’re probably right.

    If it’s not the opposite of what you thought, you know you could be wrong. Sit with that. Try to figure out why you’re wrong before you check.

    That is mindfulness. Remember to think before you search. Eventually, you will find out if you’re wrong, and in time you’ll know when you’re right most of the time without checking.

    Then don’t trust it and check anyway. Fuck it, you have agency. You do whatever the fuck you want.

  • brucethemoose@lemmy.world
    link
    fedilink
    arrow-up
    1
    ·
    edit-2
    22 days ago
    • You need digital detox. Get the LLM apps off your phone, and maybe get something like this if its affordable: https://commodore.net/callback/

    • Sign up for university counseling. Trust me, I WISH I had done this.

    • Restrict yourself to “offline” LLMs on your laptop if you MUST use them, with all the search/agenic features turned off. This turns them into more of a utility and forces you to find the references yourself, to actually integrate blocks of code, and so on.

    Treat this like smoking or some physical addiction, as that’s what these digital addictions are.


    Its probably unpopular here, but I’d also suggest you play with a “raw” LLM with completion syntax in mikupad, with the raw formatting tokens and everything:

    https://github.com/lmg-anon/mikupad

    Among other things, it dispels the “psychological mysticism” around them, and kinda shows your brain what they really are. A weather prediction model, for text. Not “AI” you brain can’t help but anthropomorphize.


    And:

    This is not your fault.

    Tech companies spend billions designing the chat UIs to be addictive, just like social media.

    You can address it though, even if it feels like everyone around you is compelling you to use. I’m not even coming at this from the anti AI angle either: I’ve been hacking at LLMs since 2022, but what OpenAI, Google and such have done is beyond fucked up.

      • brucethemoose@lemmy.world
        link
        fedilink
        arrow-up
        2
        ·
        edit-2
        22 days ago

        One thing I should have clarified. If you play with mikupad, you have to use an LLM API (or local LLM engine) that serves the old Completions API. With logprobs.

        This is “raw” LLM access without any chat formatting, and its full output before sampling.

        If you use OpenAI, Claude, or most other big providers, you are stuck with Chat APIs without logprobs, so it kinda defeats the point.

        Honestly, I don’t know of any API providers that offer this, but some role play focused ones might. Local inference engines (which are what I use) mostly have the raw API, but you need a pretty beefy computer for this.