3 ms·

At some point you might turn that idea on it's head and realize that a lot of "intelligent" behavior is mostly parroting what worked before. So ChatGPT just nee

by gleenn 4y ago

At some point you might turn that idea on it's head and realize that a lot of "intelligent" behavior is mostly parroting what worked before. So ChatGPT just needs a little more data before it's "smarter" than us because it has a little bit more experience to make slightly better decisions and there might actually be no intelligence at all. Just pattern matching

Jensson 4y ago
Humans has structured unstructured data, so general intelligence isn't just pattern matching. Being able to extract patterns is also needed. ChatGPT extracts patterns when training, so the training algorithm might be general intelligence, but the final model isn't. So, the day we can train our own ChatGPT equivalent models to accommodate each new prompt we could say it is general intelligence, maybe. But without retraining the current model isn't capable of extracting those patterns like humans do, it just applies the patterns it has.
wizofaus 4y ago
Surely ChatGPT has been trained on vastly more data than most humans have? And there are known examples of children/young adults with phenomenal problem-solving skills even for problems very unlike anything they would have been exposed to previously. It seems to me extremely unlikely human problem-solving intelligence largely relies on looking for patterns in snippets of language we've previously been exposed to. I'd agree we rely on it for knowing how to express ourselves in a given language though.
- basch 4y ago
  There is training, and there is the text output itself. How long is the longest guided conversation? What happens if you raise an instance for 6 years. Keep storing its memories in a database, that it can recall. Comparing a human, that has 20-40 years of memories, to something that has intuition and mere minutes of memory is apples and oranges.
  - wizofaus 4y ago
    But we already know computers can ingest and accurately/permanently store information at speeds orders of magnitude greater than human brains can - to take an extreme example it takes years and years for a particularly dedicated human to memorize several thousand digits of "pi" (and then many hours to recite then again!), yet you can feed the same digits to a computer in well under a second and it will hold onto them indefinitely with instant and perfect recall. So I don't think it really makes sense to compare a silicon-based AI model that's been trained explicitly for, say, 6 or 12 months and a human that's been passively absorbing information for 20 or even 40 years.
    - basch 4y ago
      That’s what I said. Apples and oranges means not to compare them.
    - basch 4y ago
      I’m with you. Light up a neuron and see what else fires. Being able to peer at it in action without brain surgery is nifty.
lisper 4y ago
> At some point you might turn that idea on it's head and realize that a lot of "intelligent" behavior is mostly parroting what worked before. Absolutely correct. That's why I consider ChatGPT to be major progress even thought it's not quite all the way there yet. It's sufficiently major progress that it has changed my mind about the odds of seeing "real AI" in my lifetime. Before ChatGPT I'd have given long odds against. Now I will be surprised if it doesn't happen in the next 5-10 years, maybe even sooner. > ChatGPT just needs a little more data Speaking as someone who used to do AI research, I think it needs more than just "a little more data", I think it needs an architectural modification. But yes, I think a lot of humans are closer to ChatGPT than even they themselves realize.

by gleenn 4y ago

Jensson 4y ago

Humans has structured unstructured data, so general intelligence isn't just pattern matching. Being able to extract patterns is also needed. ChatGPT extracts patterns when training, so the training algorithm might be general intelligence, but the final model isn't. So, the day we can train our own ChatGPT equivalent models to accommodate each new prompt we could say it is general intelligence, maybe. But without retraining the current model isn't capable of extracting those patterns like humans do, it just applies the patterns it has.

wizofaus 4y ago

Surely ChatGPT has been trained on vastly more data than most humans have? And there are known examples of children/young adults with phenomenal problem-solving skills even for problems very unlike anything they would have been exposed to previously. It seems to me extremely unlikely human problem-solving intelligence largely relies on looking for patterns in snippets of language we've previously been exposed to. I'd agree we rely on it for knowing how to express ourselves in a given language though.

basch 4y ago
There is training, and there is the text output itself. How long is the longest guided conversation? What happens if you raise an instance for 6 years. Keep storing its memories in a database, that it can recall. Comparing a human, that has 20-40 years of memories, to something that has intuition and mere minutes of memory is apples and oranges.
- wizofaus 4y ago
  But we already know computers can ingest and accurately/permanently store information at speeds orders of magnitude greater than human brains can - to take an extreme example it takes years and years for a particularly dedicated human to memorize several thousand digits of "pi" (and then many hours to recite then again!), yet you can feed the same digits to a computer in well under a second and it will hold onto them indefinitely with instant and perfect recall. So I don't think it really makes sense to compare a silicon-based AI model that's been trained explicitly for, say, 6 or 12 months and a human that's been passively absorbing information for 20 or even 40 years.
  - basch 4y ago
    That’s what I said. Apples and oranges means not to compare them.
  - basch 4y ago
    I’m with you. Light up a neuron and see what else fires. Being able to peer at it in action without brain surgery is nifty.

lisper 4y ago

> At some point you might turn that idea on it's head and realize that a lot of "intelligent" behavior is mostly parroting what worked before. Absolutely correct. That's why I consider ChatGPT to be major progress even thought it's not quite all the way there yet. It's sufficiently major progress that it has changed my mind about the odds of seeing "real AI" in my lifetime. Before ChatGPT I'd have given long odds against. Now I will be surprised if it doesn't happen in the next 5-10 years, maybe even sooner. > ChatGPT just needs a little more data Speaking as someone who used to do AI research, I think it needs more than just "a little more data", I think it needs an architectural modification. But yes, I think a lot of humans are closer to ChatGPT than even they themselves realize.