Just a guy shilling for gun ownership, tech privacy, and trans rights.

I’m open for chats on mastodon https://hachyderm.io/

my blog: thinkstoomuch.net

My email: nags@thinkstoomuch.net

Always looking for penpals!

  • 5 Posts
  • 143 Comments
Joined 2年前
cake
Cake day: 2023年12月21日

help-circle


  • I learned something interesting from my AI researcher friend.

    ChatGPT is actually pretty good at giving mundane medical advice.

    Like “I’m pretty sure I have the flu, what should I do?” Kinda advice

    His group was generating a bunch of these sorta low stakes urgent care/free clinic type questions and in nearly every scenario, ChatGPT 4 gave good advice that surveyed medical professionals agreed they would have given.

    There were some issues though.

    For instance it responded to

    “Help my toddler has the flu. How do I keep it from spreading to the rest of my family?”

    And it said

    “You should completely isolate the child. Absolutely no contact with him.”

    Which you obviously can’t do, but it is technically a correct answer.

    Better still, it was also good at knowing its limits and anything that needed more than OTC and bedrest was seemingly recognized and it would suggest going to an urgent care or ER

    So they switched to Claude and Deepseek because they wanted to research how to mitigate failures and GPT wasn’t failing often enough.





  • A few months ago now, Arizona? Arkansas maybe? Some state legalized “AI powered” home schooling systems. But it was mostly clickbait and the system is less like ChatGPT and more like the YouTube Algorithm machine learning. It takes into account the stuff that students do well at and let’s them advance beyond “grade level” limitations while also learning how to present problem areas in ways the student responds to.

    I had asked my home schooled AI researcher buddy his thoughts and he obviously liked it. I like the idea too, but my hang up was on socializing kids. That to me is the more important role of schools.

    I wouldn’t trust an LLM in this set up though. A human tutor would still need to step in for questions outside of a FAQ IMO. I love working with an LLM by giving it all the manuals, guides, and config files I used then asking where I went wrong because it can usually give me a good enough interpretation to see where to go next. But that’s just a rubber duck. My mind and skills are developed. A kid learning math for Tue first time can’t do that.











  • From what I understand its not as fast as a consumer Nvdia card but but close.

    And you can have much more “Vram” because they do unified memory. I think the max is 75% of total system memory goes to the GPU. So a top spec Mac mini M4 Pro with 48GB of Ram would have 32gb dedicated to GPU/NPU tasks for $2000

    Compare that to JUST a 5090 32GB for $2000 MSRP and its pretty compelling.

    $200 and its the 64GB model with 2x 4090’s amounts of Vram.

    Its certainly better than the AMD AI experience and its the best price for getting into AI stuff so says nerds with more money and experience than me.


  • From what I understand its not as fast as a consumer Nvdia card but but close.

    And you can have much more “Vram” because they do unified memory. I think the max is 75% of total system memory goes to the GPU. So a top spec Mac mini M4 Pro with 48GB of Ram would have 32gb dedicated to GPU/NPU tasks for $2000

    Compare that to JUST a 5090 32GB for $2000 MSRP and its pretty compelling.

    $200 and its the 64GB model with 2x 4090’s amounts of Vram.

    Its certainly better than the AMD AI experience and its the best price for getting into AI stuff so says nerds with more money and experience than me.