My AI skeptic friends are all nuts

I was recently "pair-vibe-coding" with someone who's much 'smarter' than I am, certainly when it comes to coding, especially Python. He's always been in the LLM skeptic camp, and it was fascinating to see that because of his extensive Python knowledge, his prompting was actually very minimal and weak, one might even say 'lazy'. The output he got from o3 for our given task was therefore really mediocre, and had a few hallucinations (which could've been avoided if he spent a few more seconds or minutes prompting).

I, knowing far less than him, would've had a much more elaborate prompt, and o3 would've proved a lot more competent/capable. Yet with my friend, since he knows so much already, and has such a high bar, he thinks the AI should be able to do a lot more with just a few basic words in a prompt... yet, for that same reason, he (understandably) doubts the inevitable sub-par output.

That's what makes all these debates about "Why are smart people doubting LLMs??" so pointless. The smarter you are, the less help you need, so the less prompting you do, the less context the model has, the less impressive the output, and the more the smart person thinks LLMs suck. With this logic, of course the smartest people are also the biggest skeptics!

I doubt this holds true generally. The smart coders I know who are also LLM users generally develop a decent intuition for what the models are good and bad at, and how to steer them into good performance.

It's like the problem of half-full vs half-empty. We see LLM can handle certain tasks, but counter-examples are too far from being rare. So, "LLM can do A" is always followed by "LLM fails at A".

The problem is that LLM never performs consistently. It works when it works. It doesn't when It doesn't. No one knows exactly why, and no one can tell when it's gonna fail. For example, even to this day, GPT sometimes gives me wrong calculations, even when it is instructed to use calculator for that. Who knows why it ignores the instruction, nor why it can't reliably perform the addition of two integers. That really screws up with the automation.

Anyways, I'm really tired of skeptic-skeptics. I hate some people believe "half-full" is genuinely better than "half-empty". I refuse that idea completely. It's more about which context you're in. If you need exactly a cup of water, it's half-empty. If you are lucky to have some water, it's half-full. If you have a clear direction you want to go, you discover a lot of deficiencies. If you're there just enjoying the moment, yeah, you can just keep enjoying the new toy.