So if you give a human and a system 10 tasks and the human completes 3 correctly, 5 incorrectly and 3 it failed to complete altogether… And then you give those 10 tasks to the software and it does 9 correctly and 1 it fails to complete, what does that mean. In general I’d say the tasks need to be defined, as I can give very many tasks to people right now that language models can solve that they can’t, but language models to me aren’t “AGI” in my opinion.

I forget people look at Lemmy on desktop/laptops. I just assume everyone has 15 minutes to kill and picks up there phone and opens the app they prefer, that they put on the second screen, 3rd row, 3rd column from the bottom where it belongs. If you have it somewhere else… Well maybe you have 5 columns instead of 4, or are wrong. But never on the first page … that would be obsessive. It has to hide, lurking on that second page laying in wait. For its 15 mins to shine