As a random internet user, I want to remind you, are we sure even if humans are that intelligent to begin with? All those steps you give, are not needed for intelligence.

We keep moving the goal post for what intelligence is, and last I saw we have started to divide intelligence into different categories.

LLMs are just “imitate as closely as possible human responses” for good and for bad. And now we are trying to fix that to be as right as possible, when the flaw is that we as humans are mostly always wrong.

Not OpenAI, please! Do not attack the servers! Save the ram first!