[jazzyjackson]

The output of any model is essentially random and whether it is useful or impressive is a coin flip. While most people get a mix of heads and tails, there are a few people at any time that are getting streaks of one head after another or vice versa.

So my perception is this leads to people who have good luck and perceive LLMs as near AGI because it arrives at a useful answer more often than not, and these people cannot believe there are others who have bad luck and get worthless output from their LLM, like someone at a roulette table exhorting "have you tried betting it all on black? worked for me!"