3 ms·

Lovely old science fiction story of a young man, taken from the classroom on his one day of 'school' at 18 when everybody gets 'taped' with their job knowledge.

Lovely old science fiction story of a young man, taken from the classroom on his one day of 'school' at 18 when everybody gets 'taped' with their job knowledge. They say he's incompatible with the machines, and will have to become educated the old way! He spends a life reading textbooks, studying with others like him and watching his cohort succeed and fail with their taped knowledge. Ultimately it occurs to him, who wrote those textbooks? It was people like him. The people creating the training data.

mahathu 4y ago
This would make a nice coming of age story but then the taping in question would have to occur some time during the character's childhood.
gommm 4y ago
That's Profession by Asimov Great novella (which I discovered on HN many years ago :)) https://www.abelard.org/asimov.php https://www.abelard.org/asimov.php
meh8881 4y ago
Seems unlikely to me. All this science fiction that believes the human spirit is better at things. It’s good storytelling but is it wise? Probably not. Tech gets better. Textbooks generally do not. And textbooks generally aren’t that great as they are now.
- crabbone 4y ago
  The original AI had its goal to understand and improve (make more systematic and predictable) the process of human thought / discovery. Modern day AI tries to imitate humans w/o comprehension. In some extreme cases AI researchers believe that imitation is all there is... but looking at how modern AI works, it definitely lacks the same ability we ascribe to human comprehension. So, even if there's only imitation, AI still isn't as good at it as humans are.
  - meh8881 4y ago
    I find it pretty ridiculous that people are arguing these models act without comprehension. Sentience is a whole can of worms and frankly a useless undefinable term imo. But comprehension? Please. These models comprehend concepts just fine.
    - ifdefdebug 4y ago
      By what definition of comprehension would that be? I agree they are somewhat able to transmit knowledge though.
      - CRConrad 4y ago
        So is any fiber optic cable, so “AI” is at least as smart as that.
    - crabbone 4y ago
      Daniel Dennett is a philosopher who made it sort of his favorite subject: competence without comprehension. I won't attempt to explain this better than he does. But the general idea is that these models are more elaborate versions of ELIZA, while they aren't in principle different. They work like my child who when I quiz him after reading him a page of a bed-time story on the contents of it will try to come up with plausible answers, but I can tell when he was actually paying attention / knew the words or weren't / didn't. Now, this is something I don't know whether Dennett would claim, so, it's my thought, not his. I believe that a fundamental difference is in motivation / value judgement. I.e. humans do things for a reason that they constructed based on what they perceive to be more valuable to them. Models aren't built to have values or wishes. They don't want to be able to answer more "why?" questions about whatever they "say". I mean, your reflection in the mirror is a very realistic counterfeit human being, and in many respects is indistinguishable from you, but it's nothing like a human being.
      - meh8881 4y ago
        If you request a bot to write code that can accomplish a specified task, it needs to understand the task, understand the meaning of the code, and then write code that accomplishes the task. I don’t care about appeal to sentience or what it means to “understand”. Fundamentally the meaning and of these things must be embedded in a way that leads to a synthesis of new code that solves the problem. Comparisons to ELIZA are naive. Too many philosophers waste time debating the definitions of words that they assert must represent intrinsic concepts. But they’re wrong. Sentience is just a word. It’s not well defined. You haven’t accomplished anything by finding more examples where its lack of definition is made more clear. You could say the same about comprehensions but on some level the pedantry feels so obvious and unnecessary that it’s just annoying to hear. Ffs. These modes clearly embed a comprehension of many things.

mahathu 4y ago

This would make a nice coming of age story but then the taping in question would have to occur some time during the character's childhood.

gommm 4y ago

That's Profession by Asimov Great novella (which I discovered on HN many years ago :)) https://www.abelard.org/asimov.php https://www.abelard.org/asimov.php

meh8881 4y ago

Seems unlikely to me. All this science fiction that believes the human spirit is better at things. It’s good storytelling but is it wise? Probably not. Tech gets better. Textbooks generally do not. And textbooks generally aren’t that great as they are now.

crabbone 4y ago
The original AI had its goal to understand and improve (make more systematic and predictable) the process of human thought / discovery. Modern day AI tries to imitate humans w/o comprehension. In some extreme cases AI researchers believe that imitation is all there is... but looking at how modern AI works, it definitely lacks the same ability we ascribe to human comprehension. So, even if there's only imitation, AI still isn't as good at it as humans are.
- meh8881 4y ago
  I find it pretty ridiculous that people are arguing these models act without comprehension. Sentience is a whole can of worms and frankly a useless undefinable term imo. But comprehension? Please. These models comprehend concepts just fine.
  - ifdefdebug 4y ago
    By what definition of comprehension would that be? I agree they are somewhat able to transmit knowledge though.
    - CRConrad 4y ago
      So is any fiber optic cable, so “AI” is at least as smart as that.
  - crabbone 4y ago
    Daniel Dennett is a philosopher who made it sort of his favorite subject: competence without comprehension. I won't attempt to explain this better than he does. But the general idea is that these models are more elaborate versions of ELIZA, while they aren't in principle different. They work like my child who when I quiz him after reading him a page of a bed-time story on the contents of it will try to come up with plausible answers, but I can tell when he was actually paying attention / knew the words or weren't / didn't. Now, this is something I don't know whether Dennett would claim, so, it's my thought, not his. I believe that a fundamental difference is in motivation / value judgement. I.e. humans do things for a reason that they constructed based on what they perceive to be more valuable to them. Models aren't built to have values or wishes. They don't want to be able to answer more "why?" questions about whatever they "say". I mean, your reflection in the mirror is a very realistic counterfeit human being, and in many respects is indistinguishable from you, but it's nothing like a human being.
    - meh8881 4y ago
      If you request a bot to write code that can accomplish a specified task, it needs to understand the task, understand the meaning of the code, and then write code that accomplishes the task. I don’t care about appeal to sentience or what it means to “understand”. Fundamentally the meaning and of these things must be embedded in a way that leads to a synthesis of new code that solves the problem. Comparisons to ELIZA are naive. Too many philosophers waste time debating the definitions of words that they assert must represent intrinsic concepts. But they’re wrong. Sentience is just a word. It’s not well defined. You haven’t accomplished anything by finding more examples where its lack of definition is made more clear. You could say the same about comprehensions but on some level the pedantry feels so obvious and unnecessary that it’s just annoying to hear. Ffs. These modes clearly embed a comprehension of many things.