I have fleas. https://www.snand.org/

  • 0 Posts
  • 21 Comments
Joined 3 years ago
cake
Cake day: June 28th, 2023

help-circle



  • 4grams@awful.systemstoFuck AI@lemmy.worldWhy do you hate AI?
    link
    fedilink
    English
    arrow-up
    5
    arrow-down
    3
    ·
    edit-2
    6 days ago

    I don’t hate AI, I kinda like it and I do think it will redefine computing in the future.

    I hate the people behind it, and who are pushing it. I hate the hype, I hate the pressure, I hate how dangerous it is without guardrails, I hate the stupidity of people using it, and getting addicted to it because it’s an ego stoking machine. I hate the entire industry around it.

    But I really enjoy using and learning about it in my own sandbox. I use a number of local LLMs successfully for research and learning. But I don’t trust them, at all, I think of them as an egotistical knowitall who has no problem lying to make themselves feel smart. There’s tons of useful info to get from them, but you have to understand what you are dealing with.

    I’m very curious what the future looks like. It really depends on us though. Critical thinking and being observant are the new critical skills for success in the he future. Unfortunately neither are particularly common these days so I have a feeling I will continue to hate for a while now…








  • Yeah, it’s a huge disaster waiting to happen. I am playing with it because it is some interesting learning but completely sandboxed, using a local LLM, and I have given it no personal information. It’s really cool to see it work, to watch the communication between it and the LLM.

    But seriously, I’d never, ever attempt playing with it without my decades of infrastructure experience. I’d never give it public access nor any personal information. Still, I am encouraged that it’s functional enough in an entirely self hosted stack to be able to learn and play. I still think the future of AI should be local and openclaw, despite its fundamental issues is the first time I’ve played with AI and could conceive a future use case.

    In the end, I think the bubble will pop, I just hope the other side has something of value to come out of all this insanity. As brick stupid as openclaw is, it’s at least enough of a proof of concept to keep a sliver of hope that there may be.


  • I am playing with it, sandboxed in an isolated environment, only interacting with a local LLM and only connected to one public service with a burner account. I haven’t even given it any personal info, not even my name.

    It’s super fascinating and fun, but holy shit the danger is outrageous. Multiple occasions, it’s misunderstood what I’ve asked and it will fuck around with its own config files and such. I’ve asked it to do something and the result was essentially suicide as it ate its own settings. I’ve only been running it for like a week but have had to wipe and rebuild twice already (probably could have fixed it, but that’s what a sandbox is for). I can’t imagine setting it loose on anything important right now.

    But it is undeniably cool, and watching the system communicate with the LLM model has been a huge learning opportunity.






  • It’s a lot of fun, but not an internet replacement yet. They did just come out with a new dual band that has enough bandwidth to do something useful. Can’t wait to give it a try.

    Still very worth setting up. I have a node that runs in my attic, and a few that I can take around with me. I can get good single most anywhere in my neighborhood, and I have enough nodes nearby that I can pretty much communicate with people all over the metro area (I’m in the twin cities).