2 ms·
Well, yeah… turns out that goal wasn’t a good indicator for AGI, so we re-evaluated. That’s changing your hypothesis in the face of evidence, not “moving the go
by emp17344 6mo ago
Well, yeah… turns out that goal wasn’t a good indicator for AGI, so we re-evaluated. That’s changing your hypothesis in the face of evidence, not “moving the goalposts” in the fallacious sense.
- Mistletoe 6mo agoWhat’s the indicator for AGI now? We are so far past the Turing Test it isn’t funny. In fact the models now are too intelligent, you would never think a human would have that much knowledge quickly about a subject you chose at random.