We keep having the discussion about what is AGI and keep changing the benchmarks, but there is a fundamental problem. The problem is defined.
In the real world problems are not defined, you can yap every coded app human or ai into newsworthiness, but what actually decides the success is the money it makes.
We might have openclaw and some other vibe coded examples, but they have been steered by humans. Also on the flip side we have millions on token spend on apps that will never make it back.
I’m aware that there are a lot of nuances, but my current view is that AI is really good in teaching you foreign techniques frameworks coding languages but once you really good at it, it’s faster to build it yourself.
So I’m questioning a lot of things rn and I’m happy to discuss, but the most prominent thought in my mind right now is.
How much better is a human in making money then AI given the same rights and input (computer)?
As an entrepreneur working on the wrong problems can be costly and whilst even at expert tasks AI excels I wonder how it performs on vague goals?
e.g.: improve my software
No comments yet.
