4 ms·

Philosophers have been debating what "meaning" means for a while. But in the context of ChatGPT and LLMs, when people talk about these system "not understanding

Philosophers have been debating what "meaning" means for a while. But in the context of ChatGPT and LLMs, when people talk about these system "not understanding" a given idea, they point to the things handling the properties of words but not consistently. -- In your example, ChatGPT might translate each of your conditions into a rule and use something like an algorithm to solve the problem. Or it could notice the similarity of that problem to similar problems and interpolate an answer like the answer it's seen to those problems. But this can easily result in answers suddenly changing.

scarface74 4y ago
Another example: “I’m thinking about getting either the Amex Hilton Aspire card that gives me 150,000 Hilton Honors points for signing up and has a $475 AF or the Amex Gold that gives me 70,000 Membership Reward points for signing up and has a $250 fee. A Hilton Honors point is worth .7 cents. A membership reward point is worth 1.4 cents. Which one would give me a better return in the first year” ChatGPTs answer: To determine which card would give you a better return in the first year, you'll need to consider the total value of the rewards earned, as well as the cost of the annual fee. For the Hilton Aspire card, the 150,000 Hilton Honors points are worth $1,050 (150,000 x 0.7 cents per point). However, the $475 annual fee would offset a significant portion of that value. For the Amex Gold card, the 70,000 Membership Reward points are worth $980 (70,000 x 1.4 cents per point). The $250 annual fee is lower than that of the Hilton Aspire card, but it still needs to be factored into the overall value. Assuming you maximize the value of your rewards by redeeming them for their highest value and using all the card's benefits, here's a comparison of the two cards' net value in the first year: Hilton Aspire: $1,050 (value of points earned) - $475 (annual fee) = $575 net value Amex Gold: $980 (value of points earned) - $250 (annual fee) = $730 net value Therefore, based on this analysis, the Amex Gold card would give you a better return in the first year. However, keep in mind that the value of the rewards can vary depending on how you redeem them, and your personal spending habits and travel preferences may also play a role in determining which card is right for you.
- joe_the_user 4y ago
  The complex behavior you're showing doesn't prove what you think it proves - it still doesn't show it's using the consistent rules that a person would expect. But it does show that people extrapolate complex behavior to "understanding" in the way humans do, which machines generally don't.
  - scarface74 4y ago
    I’m just trying to “prove” that it isn’t just randomly statistically choosing the next logical word. It has to know context and have some level of “understanding” of other contexts. People are acting as if ChatGPT is a glorified Eliza clone.
    - sramsay 4y ago
      This. We have had a lot people -- including journalists and academics with big microphones -- learn for the first time what a Markov chain is, and then conclude that ChatGPT is a "just Markov chains" (or whatever similarly reductive concept). They really, really don't know what they're talking about it, and yet it's becoming a kind of truth through repetition. Pretty soon, the bots will start saying it!
      - _2uwr 4y ago
        > They really, really don't know what they're talking about it, and yet it's becoming a kind of truth through repetition. Kind of like religion or that people working for the state are more trustworthy than people taking drugs or sleeping on the street or under the age of 18. >Pretty soon, the bots will start saying it! We are chemical based repetition machines, psychologists see this with kids using bobo dolls exposed to new ideas on tv or in books repeating learned behaviour on bobo dolls. I think some of the chemicals we make like https://en.wikipedia.org/wiki/N,N-Dimethyltryptamine https://en.wikipedia.org/wiki/N,N-Dimethyltryptamine actually help to create new idea's, as many people say they come up with solutions after some sleep. There appears to be a sub culture in silicon valley were microdosing lsd helps to maintain the creativity with coding. It would seem logical for the bots to start saying it. If the bots start amplifying flawed knowledge like a lot of Reddit content or Facebook content, the internet will need to deal with the corruption of the internet, like using Wikipedia as a source of reference. https://en.wikipedia.org/wiki/Wikipedia:List_of_hoaxes_on_Wikipedia/Less_than_one_year https://en.wikipedia.org/wiki/Wikipedia:List_of_hoaxes_on_Wi... https://en.wikipedia.org/wiki/Reliability_of_Wikipedia https://en.wikipedia.org/wiki/Reliability_of_Wikipedia https://en.wikipedia.org/wiki/List_of_Wikipedia_controversies https://en.wikipedia.org/wiki/List_of_Wikipedia_controversie... In practice, sources of reference will need cleaning up and the question about who should be responsible for that will probably ensue. It will call into question such entities like encyclopaedia companies who have been the gatekeeper of the knowledge they sell, in much the same way the Guinness World Record are the self declared arbitrators of world records. We already see Judges working in a semi cohesive manner, but the appeals process tells you its flawed and inefficient, ergo can the judiciary even be trusted to interpret law properly? Its also interesting how the financial system also reinforces this mob rule mentality and not always for the better, but then some people will no doubt suggest we should be grateful we have bungled our way to this point in history.

scarface74 4y ago

Another example: “I’m thinking about getting either the Amex Hilton Aspire card that gives me 150,000 Hilton Honors points for signing up and has a $475 AF or the Amex Gold that gives me 70,000 Membership Reward points for signing up and has a $250 fee. A Hilton Honors point is worth .7 cents. A membership reward point is worth 1.4 cents. Which one would give me a better return in the first year” ChatGPTs answer: To determine which card would give you a better return in the first year, you'll need to consider the total value of the rewards earned, as well as the cost of the annual fee. For the Hilton Aspire card, the 150,000 Hilton Honors points are worth $1,050 (150,000 x 0.7 cents per point). However, the $475 annual fee would offset a significant portion of that value. For the Amex Gold card, the 70,000 Membership Reward points are worth $980 (70,000 x 1.4 cents per point). The $250 annual fee is lower than that of the Hilton Aspire card, but it still needs to be factored into the overall value. Assuming you maximize the value of your rewards by redeeming them for their highest value and using all the card's benefits, here's a comparison of the two cards' net value in the first year: Hilton Aspire: $1,050 (value of points earned) - $475 (annual fee) = $575 net value Amex Gold: $980 (value of points earned) - $250 (annual fee) = $730 net value Therefore, based on this analysis, the Amex Gold card would give you a better return in the first year. However, keep in mind that the value of the rewards can vary depending on how you redeem them, and your personal spending habits and travel preferences may also play a role in determining which card is right for you.

joe_the_user 4y ago
The complex behavior you're showing doesn't prove what you think it proves - it still doesn't show it's using the consistent rules that a person would expect. But it does show that people extrapolate complex behavior to "understanding" in the way humans do, which machines generally don't.
- scarface74 4y ago
  I’m just trying to “prove” that it isn’t just randomly statistically choosing the next logical word. It has to know context and have some level of “understanding” of other contexts. People are acting as if ChatGPT is a glorified Eliza clone.
  - sramsay 4y ago
    This. We have had a lot people -- including journalists and academics with big microphones -- learn for the first time what a Markov chain is, and then conclude that ChatGPT is a "just Markov chains" (or whatever similarly reductive concept). They really, really don't know what they're talking about it, and yet it's becoming a kind of truth through repetition. Pretty soon, the bots will start saying it!
    - _2uwr 4y ago
      > They really, really don't know what they're talking about it, and yet it's becoming a kind of truth through repetition. Kind of like religion or that people working for the state are more trustworthy than people taking drugs or sleeping on the street or under the age of 18. >Pretty soon, the bots will start saying it! We are chemical based repetition machines, psychologists see this with kids using bobo dolls exposed to new ideas on tv or in books repeating learned behaviour on bobo dolls. I think some of the chemicals we make like https://en.wikipedia.org/wiki/N,N-Dimethyltryptamine https://en.wikipedia.org/wiki/N,N-Dimethyltryptamine actually help to create new idea's, as many people say they come up with solutions after some sleep. There appears to be a sub culture in silicon valley were microdosing lsd helps to maintain the creativity with coding. It would seem logical for the bots to start saying it. If the bots start amplifying flawed knowledge like a lot of Reddit content or Facebook content, the internet will need to deal with the corruption of the internet, like using Wikipedia as a source of reference. https://en.wikipedia.org/wiki/Wikipedia:List_of_hoaxes_on_Wikipedia/Less_than_one_year https://en.wikipedia.org/wiki/Wikipedia:List_of_hoaxes_on_Wi... https://en.wikipedia.org/wiki/Reliability_of_Wikipedia https://en.wikipedia.org/wiki/Reliability_of_Wikipedia https://en.wikipedia.org/wiki/List_of_Wikipedia_controversies https://en.wikipedia.org/wiki/List_of_Wikipedia_controversie... In practice, sources of reference will need cleaning up and the question about who should be responsible for that will probably ensue. It will call into question such entities like encyclopaedia companies who have been the gatekeeper of the knowledge they sell, in much the same way the Guinness World Record are the self declared arbitrators of world records. We already see Judges working in a semi cohesive manner, but the appeals process tells you its flawed and inefficient, ergo can the judiciary even be trusted to interpret law properly? Its also interesting how the financial system also reinforces this mob rule mentality and not always for the better, but then some people will no doubt suggest we should be grateful we have bungled our way to this point in history.