4 ms·

That headline really resonates as I spent many years in Las Vegas. LLMs in general have very slot machine like qualities. WHen they “pay off” they are great.

astrobe_ 1y ago
.. But the house always wins ?
- ChuckMcM 1y ago
  Right? I hope that at some point the weights system will become clear enough that one could bias against wrong answers. Just “telling” it that is wrong biases for the current token cache and doesn’t back propagate to the model. Of course back propagating user input would be very exploitable so I don’t see that happening.
  - prmph 1y ago
    What about if further individualized training could occur at the user site?
- marcosdumay 1y ago
  Quite possibly. They are subsidized right now, but those things are really expensive.
svantana 1y ago
Oh great, together we are two datapoints that reinforce the OP's hypothesis. I have a strong distaste for gambling in all forms and I don't enjoy using opaque, unreliable text generators either.
leptons 1y ago
There are many ways to describe LLMs and coding assistants. The slot machine is a pretty good one. I also compare "AI" to using a Ouija Board. It's not meant for getting real answers or truth. The game is to select the next letter in a word to create a sequence of words. It's an entertainment prop, and LLMs should be treated similarly. I have also compared "Artificial Intelligence" to artificial flavors. Beaver anus is used as artificial vanilla flavoring (that is a real thing that happens), and "AI"/LLMs are the digital equivalent. Real vanilla tastes so much better, even if it is more expensive. I have no doubt that code written by real humans works much better than AI slop code. Having tried the various "AI" coding assistants, and I am not at all impressed with what it creates. Half the time if I ask for "vanilla", it will tell me "banana".

astrobe_ 1y ago

.. But the house always wins ?

ChuckMcM 1y ago
Right? I hope that at some point the weights system will become clear enough that one could bias against wrong answers. Just “telling” it that is wrong biases for the current token cache and doesn’t back propagate to the model. Of course back propagating user input would be very exploitable so I don’t see that happening.
- prmph 1y ago
  What about if further individualized training could occur at the user site?
marcosdumay 1y ago
Quite possibly. They are subsidized right now, but those things are really expensive.

svantana 1y ago

Oh great, together we are two datapoints that reinforce the OP's hypothesis. I have a strong distaste for gambling in all forms and I don't enjoy using opaque, unreliable text generators either.

leptons 1y ago

There are many ways to describe LLMs and coding assistants. The slot machine is a pretty good one. I also compare "AI" to using a Ouija Board. It's not meant for getting real answers or truth. The game is to select the next letter in a word to create a sequence of words. It's an entertainment prop, and LLMs should be treated similarly. I have also compared "Artificial Intelligence" to artificial flavors. Beaver anus is used as artificial vanilla flavoring (that is a real thing that happens), and "AI"/LLMs are the digital equivalent. Real vanilla tastes so much better, even if it is more expensive. I have no doubt that code written by real humans works much better than AI slop code. Having tried the various "AI" coding assistants, and I am not at all impressed with what it creates. Half the time if I ask for "vanilla", it will tell me "banana".