[chankstein38]

My problem is I feel like my irritation is with things that are out of my control. People not caring about quality when they produce things that I spend good money on (and have no abilities to turn into a new manufacturing startup or something). Software (Adobe, Blender, Davinci Resolve, etc) that has a feature that doesn't feel like it makes sense or something that just doesn't seem to work. Another thing that I have no real knowledge about. I'm a programmer but I don't think there's a good reason for me to write my own 3d software or image/video editor.

Side note, I've never looked at this person's "AI Powered Podcast" but how do they even know they're getting real information about Indian History? I was talking to ChatGPT last night about whether or not it was possible to adjust the level of the water in a toilet bowl and it confidently told me I could and how to do it. I kept discussing it with it and, at one point, it slipped in a comment about how it's controlled by the level of the trap in the back of the toilet. I said "So it's NOT possible and you've just wasted my time?" and it was like "Yep, sorry about that"

Same night I asked it something and provided it a link to read and I asked a question about it. It indicated it had gone to the page and, I assume, consumed the text on the page. It answered the question but then said something that made me realize it might be wrong. I asked it to tell me where it saw that and it told me it wasn't actually on the page.

I ask because it sounds interesting to make an "AI Powered Podcast" about things I'm interested in. I frequently want to listen to a podcast while mowing my grass or something and struggle to find interesting, long-form things about the specific topics that I'm thinking about. (Radio astronomy, weather, astronomy, music, whatever) So I'd love to do this but I trust AI outputs so little...

[nartho]

>but how do they even know they're getting real information about Indian History

They do not. AI and LLMs are awesome tools when they're being used correctly, with their limitations in mind, but for a big majority of people who don't have a clue about how they work, it's going to be a disaster. We will see more and more of this garbage because it's easy money. AI generated podcasts, youtube videos etc, until people start believing hallucinated junk.

[Eddy_Viscosity2]

I'll add to the list of irritations that are out of my control, managers who only value following process instead of achieving outcomes. Even when it is very very clear and foreseeable that a particularly problem should use a different process because of its specific context and constraints, they just don't and won't. Deviating from established process is a greater sin than waste and failure.

[j4mehta]

Yeah fair point! Honestly, the low faith in AI is probably a good stance to have right now.

The AI podcast was an experiment for me, I created it by feeding reference content (textbooks, primary material, etc) to NotebookLM. The results were _okay_, some episodes were good but the discussion was far too surface level, and I couldn't direct it towards any specific direction.

The tech is evolving so fast though, so I do think there's an opportunity in that space to build something interesting.