Comment by Fraterkes

Comment by Fraterkes 7 hours ago

1 reply

The METR graph proposes a 6 year trend, based largely on 4 datapoints before 2024. I get that it is hard to do analyses since were in uncharted territory, and I personally find a lot of the AI stuff impressive, but this just doesn't strike me as great statistics.

benlivengood 6 hours ago

I agree that we don't have any good statistical models for this. If AI development were that predictable we'd likely already be past a singularity of some sort or in a very long winter just by reverse-engineering what makes the statistical model tick.