3 ms·

I think "hasn't seen before" is a bit of an overstatement. Sure, the problem is new in the literal sense that it does exist verbatim elsewhere, but arguably, an

I think "hasn't seen before" is a bit of an overstatement. Sure, the problem is new in the literal sense that it does exist verbatim elsewhere, but arguably, any competition problem is hardly novel: they are all some permutation of problems that exist and have been solved before: pathfinding, optimization, etc. I don't think anyone is pretending to break new scientific ground in 5 hours.

chpatrick 1y ago
It's not new scientific ground but a machine beating a challenging computer science problem unassisted is a big deal. If they can do that then there are a lot of other challenging things they can do.
- raspasov 1y ago
  Like what exactly? As far as I can tell, the drug discovery is fizzling out, so it's not talked about much. Toxicity, for one, is a big problem, and the AI is not going to tell you whether the new drug it just concocted is suitable for humans or not. Small model solves an easy problem; big model solves a challenging problem. I wouldn't call those problems; they are more like invented puzzles. Perfect match for the AI marketing department to "solve".
  - chpatrick 1y ago
    Whatever you say.
    - raspasov 1y ago
      Have an AI model prove that P is not equal to NP. Then I’ll be impressed.
      - chpatrick 1y ago
        So you'll only be impressed if it's superintelligent and solves a problem no human has? That's a ridiculously high bar.
        raspasov 1y ago
        Humans have been shown to solve problems and discover things that "no human has" up to that point. So I wouldn't even call it "superintelligent". But it would definitely be truly useful at that point! The marketing speak of some companies goes into effectively fantastical territory. Some claims that are made effectively imply P = NP, which is like bad TV-series-level science fiction, but some people fall for it.

3 ms·

I think "hasn't seen before" is a bit of an overstatement. Sure, the problem is new in the literal sense that it does exist verbatim elsewhere, but arguably, an

by raspasov 1y ago

chpatrick 1y ago
It's not new scientific ground but a machine beating a challenging computer science problem unassisted is a big deal. If they can do that then there are a lot of other challenging things they can do.
- raspasov 1y ago
  Like what exactly? As far as I can tell, the drug discovery is fizzling out, so it's not talked about much. Toxicity, for one, is a big problem, and the AI is not going to tell you whether the new drug it just concocted is suitable for humans or not. Small model solves an easy problem; big model solves a challenging problem. I wouldn't call those problems; they are more like invented puzzles. Perfect match for the AI marketing department to "solve".
  - chpatrick 1y ago
    Whatever you say.
    - raspasov 1y ago
      Have an AI model prove that P is not equal to NP. Then I’ll be impressed.
      - chpatrick 1y ago
        So you'll only be impressed if it's superintelligent and solves a problem no human has? That's a ridiculously high bar.
        raspasov 1y ago
        Humans have been shown to solve problems and discover things that "no human has" up to that point. So I wouldn't even call it "superintelligent". But it would definitely be truly useful at that point! The marketing speak of some companies goes into effectively fantastical territory. Some claims that are made effectively imply P = NP, which is like bad TV-series-level science fiction, but some people fall for it.