5 ms·

Text is a representation of human thought. This may look like text phenomenologically, but it has no meaning except when it’s filtered by a human. There is a de

Text is a representation of human thought. This may look like text phenomenologically, but it has no meaning except when it’s filtered by a human. There is a deep ontological flaw with these applications.

dev_snd 5y ago
"Ceci n'est pas une pipe", as Magritte woyld say. It's always the observer, that makes a thing into what we believe it is. Without an observer nothing has any meaning. The AI does not have to understand words and objects in the same way as you do to have real world use cases.
- hashimotonomora 5y ago
  So you would say that The Library of Babel (Borges) as a system is AI because for an observer, some texts make sense? You can also see faces on clouds, but is that you or is it the cloud representing a face?
  - dev_snd 5y ago
    Interesting question. I guess what makes us call something an AI is that it's an interactive recombination of a corpus of knowledge as opposed to a static one (such is the library). GPT-3 is far from sentient, that's for sure, but it's more interactive than any library. People used the clouds to tell the future in the past, so there certainly was a value to reading them. Maybe GPT-3 is the next best thing to do the same, maybe it's more, time will tell.
bogdanoff_2 5y ago
What gives meaning to text? How do you know there's something more to text that the "phenomenology"?
- hashimotonomora 5y ago
  The way I see it, text is an external representation of human thought. If I could transmit my thoughts to you by telepathy, wee wouldn’t use text. Text cannot represent everything, so we invented drawings and paintings as representation of our thought. Text in itself has no meaning. The fact that a combination of letters generated by a Bayesian filter makes sense is just a coincidence. It just passes our “makes sense” filter which allow our brains to differentiate signal from noise. But it doesn’t represent anything.

dev_snd 5y ago

"Ceci n'est pas une pipe", as Magritte woyld say. It's always the observer, that makes a thing into what we believe it is. Without an observer nothing has any meaning. The AI does not have to understand words and objects in the same way as you do to have real world use cases.

hashimotonomora 5y ago
So you would say that The Library of Babel (Borges) as a system is AI because for an observer, some texts make sense? You can also see faces on clouds, but is that you or is it the cloud representing a face?
- dev_snd 5y ago
  Interesting question. I guess what makes us call something an AI is that it's an interactive recombination of a corpus of knowledge as opposed to a static one (such is the library). GPT-3 is far from sentient, that's for sure, but it's more interactive than any library. People used the clouds to tell the future in the past, so there certainly was a value to reading them. Maybe GPT-3 is the next best thing to do the same, maybe it's more, time will tell.

bogdanoff_2 5y ago

What gives meaning to text? How do you know there's something more to text that the "phenomenology"?

hashimotonomora 5y ago
The way I see it, text is an external representation of human thought. If I could transmit my thoughts to you by telepathy, wee wouldn’t use text. Text cannot represent everything, so we invented drawings and paintings as representation of our thought. Text in itself has no meaning. The fact that a combination of letters generated by a Bayesian filter makes sense is just a coincidence. It just passes our “makes sense” filter which allow our brains to differentiate signal from noise. But it doesn’t represent anything.