3 ms·

>This is a book from 2006 accurately using the salient features of LLMs popularised in the 2020s I'm a huge fan of Peter Watts' work, but to be fair that is at

>This is a book from 2006 accurately using the salient features of LLMs popularised in the 2020s I'm a huge fan of Peter Watts' work, but to be fair that is at the very least a close-cousin to the Chinese Room thought experiment that's been around for a while.

eulgro 3y ago
And so it is said in the book.
- lostmsu 3y ago
  And since the book itself is a work of fiction, it doesn't really tell us what will reality look like.
  - XorNot 3y ago
    Yes and no: good science fiction takes a trend and extrapolates. Extrapolation should be based on logic (to be good - but you can also just aim for entertaining and go wild). The details might differ, but it's worth thinking about whether the implications make sense. Watt's certainly asks a lot of good questions which other orders have also asked - i.e. Accelerando[1] is very interested in the idea of "what happens to mankind when economic systems can no longer be participated in by recognizably human minds?" - taken out to the limits of known physics. [1] https://www.antipope.org/charlie/blog-static/fiction/accelerando/accelerando-intro.html https://www.antipope.org/charlie/blog-static/fiction/acceler... ^^^ You read the full-text here, I highly recommend it.
  - gpderetta 3y ago
    Of course it doesn't, but it is a thought experimented presented as a novel.
- ineedasername 3y ago
  Yes, Blindsight is also the book that pointed me towards Metzinger’s Being No One.
  - drkleiner 3y ago
    Solid book, Metzinger is really underrated
garyNeedsa 3y ago
Yeah that’s the thing; LLMs like all software are just the latest codification of existing theories and memes, that’s exactly what they’re trained on. That’s exactly the path we envisioned to create them decades ago. The concept wasn’t new in 2006 either. Not really seeing the “now that’s sci fi!” emotion here. Most of our historical compute problems were coupled to network reliability, performance. Those bottlenecks are largely solved; oh wow explosion in AI progress. This isn’t new it’s more like the Higgs; just needed to wait for hardware.

eulgro 3y ago

And so it is said in the book.

lostmsu 3y ago
And since the book itself is a work of fiction, it doesn't really tell us what will reality look like.
- XorNot 3y ago
  Yes and no: good science fiction takes a trend and extrapolates. Extrapolation should be based on logic (to be good - but you can also just aim for entertaining and go wild). The details might differ, but it's worth thinking about whether the implications make sense. Watt's certainly asks a lot of good questions which other orders have also asked - i.e. Accelerando[1] is very interested in the idea of "what happens to mankind when economic systems can no longer be participated in by recognizably human minds?" - taken out to the limits of known physics. [1] https://www.antipope.org/charlie/blog-static/fiction/accelerando/accelerando-intro.html https://www.antipope.org/charlie/blog-static/fiction/acceler... ^^^ You read the full-text here, I highly recommend it.
- gpderetta 3y ago
  Of course it doesn't, but it is a thought experimented presented as a novel.
ineedasername 3y ago
Yes, Blindsight is also the book that pointed me towards Metzinger’s Being No One.
- drkleiner 3y ago
  Solid book, Metzinger is really underrated

garyNeedsa 3y ago

Yeah that’s the thing; LLMs like all software are just the latest codification of existing theories and memes, that’s exactly what they’re trained on. That’s exactly the path we envisioned to create them decades ago. The concept wasn’t new in 2006 either. Not really seeing the “now that’s sci fi!” emotion here. Most of our historical compute problems were coupled to network reliability, performance. Those bottlenecks are largely solved; oh wow explosion in AI progress. This isn’t new it’s more like the Higgs; just needed to wait for hardware.