When it’s pissing you off stop. Take a beat and think about whether you can go around instead of through. Or whether it’s worth going through once in a manner that makes it easier next time. Walk around the lake, or build a boat. Walk around the wood, or build an path and turn it into a road by increments.
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...
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.
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.
I think the real challenge is that what bothers most people are actually minor irritations. We don't notice the big thing shouting at us that it's a problem.
This was my experience which lead me into neurotech/sleeptech.
I'm a lifelong insomniac. I was going through sleep labs as a kid. I had done all the CBTi (though I don't think it was called that in the 80s), sleep hygiene, etc. etc. I'm quite healthy, just crap sleeper.
I was almost 50 when at 3am I thought to myself "I don't care if I sleep, I just don't want to be tired anymore".
I thought that was an interesting thought. Poor sleep had been ruining my life for the most part of 5 decades.
So I started looking into the latest in sleep research (I had worked in healthtech at Australia's science and technology research agency), and it all seemed very much the same as what I'd heard before. I started tracking everything, trying to figure out if I could build some sort of algorithm that picked up a signal of what I was doing when I had good sleep vs bad.
This was completely the wrong approach. My irritation wasn't truly that I couldn't sleep enough. That was part of it, but the bigger irritation was that I was always tired. Even though I asked the right question, the global mentality was "you need to sleep longer".
That was until I discovered research in slow-wave enhancement, which increases the synchronous firing of neurons which are the defining marker of deep sleep. This is tapping into the restorative function of sleep, not focusing on sleep time.
So I think that finding the irritation that may be hidden from you is step 1. But beyond that, who else has had the same irritation, and what have they done about it. Why did their solutions not work. Then, why do you think you can do it better?
If you're curious about what we're building, you can check out https://affectablesleep.com
PaulHoule•1d ago
firesteelrain•1d ago
PaulHoule•1d ago
I find myself irritated by
https://www.fastcompany.com/91324550/kerning-on-pope-francis...
like Italy is hotbed of art and culture and one reason you go to a Catholic Church is to appreciate a beautiful environment. How many things has the church carved in stone and done correctly? Maybe 1000 years from now the tomb will be still like that and it will remind people that we had these terrible machines called ‘computers’ and aren’t we glad they are all gone.
vakkermans•1d ago
firesteelrain•1d ago