4 ms·
> A strong AI can certainly do this. A strong AI can say something different than the thing it's about to say? That's a neat trick. Not a part of any definitio
by Bakkot 14y ago
> A strong AI can certainly do this.
A strong AI can say something different than the thing it's about to say? That's a neat trick. Not a part of any definition of strong AI I've ever heard, though. Could a human do it?
- Evbn 14y agoNo, a human could not. Similarly, No human nor computer can name all the real numbers. Luckily, it turns out that there were only ever be a finite number of questions ever asked, which require only finite answers. Similarly, the halting problem is solvable on finite memory machines, aka the only kind that exist.
- Dove 14y agoThat's not exactly what I said though. I didn't say "run yourself." What I said was, "Run this code with this input, and then do something different." A human could certainly do that. A strong AI should certainly be able to. Hence it can't be defined in terms of code and input. Not a problem for humans. My code and input are massively inaccessible. Modeling them, especially modelling all the input I've received--could involve modeling a big chunk of the universe. Whether or not they're even deterministic is an open philosophical problem. I'm pretty safe from being asked to "run" myself. Not so for a simple Strong-AI chatbot, though. If it's just a program a computer can run, if you can enumerate all the input . . . then by definition, it can't handle a simple request that any human would be able to handle. Not strong.