Sounds like a great space to dive into. Plenty of depth so you won't run out of interesting things to learn and there are lots of ways AI can help others.
If climate's your thing, you can check out https://www.climatechange.ai/. I'd bet similar communities exist around other areas for social good too. Good luck :)
I've been exploring pretty much the same questions for the last few years. The first couple years felt distressing because I didn't feel like I was making progress quickly enough and I had way more questions than answers. Lately, I feel more positive about the direction of travel, and I've made peace with this being a slow, difficult journey. Seems like it should be "easy", but it's really not. Accepting that and not feeling too hard on myself helped a lot.
I haven't figured out a Grand Unified Theory or anything, but a few ideas that seem important to me:
1. Altruism & Relationships - covered extensively here already. Cultivating a few deep relationships and picking altruistic goals seem to be the scalable approaches to escaping the Hedonistic Treadmill and they're probably necessary components of any post-financial independence life strategy.
2. Growth - I'm not sure if this is a universal human thing or just something entrepreneur types are predisposed to, but I've realised that I'm much happier when I'm learning and growing. When the learning curve flattens out, I get feel unsettled. This seems to line up with your observation that you intensely take up a hobby, then shift to something new. Startups scratch this itch well because the company's growth forces constant personal growth. Initially, I tried to transcend this need to always be in motion thinking it was unsustainable over the long term - but lately I've been more accepting of it.
3. Change the Game - the trouble with growth as you approach the peak of whatever career/goal/journey you're on is it becomes exponentially harder to level up and satisfy that need for motion. I think the solution is to pivot to a totally different journey where you can start as a beginner and have a lot of attainable learning & growth ahead of you. I suspect this is a big part of why successful people tend to pursue philanthropy at the height of their careers - it's a whole new game, with new challenges to overcome.
From those ideas, I decided to pick a difficult, new domain to focus on with a goal of figuring out how to do some meaningful good. I happened to choose climate change, but health, education, poverty reduction, democracy, or any number of other issues fit the bill of "hard enough to provide years of learning/growth + focus is helping others". By picking something and committing, I avoided getting stuck in and endless loop of trying to decide what the optimal area is - a bias to action helps a lot when you get stuck with too much choice.
For me, the concrete starting point was Googling "best climate change books", reading half a dozen of them, and trusting my curiosity to guide me from there. I've spent the last year learning, and now I'm converging on starting a startup - although there are plenty of other structures that could work too (non-profits, volunteering, working on open source, consulting, advising, etc). The journey hasn't given me total clarity or magically solved everything - but by imposing some structure ("work full-time to figure out how you can help with climate change") on my life, things are at least less confusing.
+1 for block-programming, even for older kids/teenagers. I taught a high school CS class using Snap.
At first, I thought it was silly to teach "drag & drop" programming and would have preferred a high-level scripting language. But, after teaching 2 semesters of Snap and observing classes in Java, I'm a huge fan.
2 main reasons:
1. Time-to-cool-result is really low. The first project was a Mario-like game. Less than 10 hours from "never programmed" to "built Mario game" is a really powerful first experience. (Demo: http://snap.berkeley.edu/snapsource/snap.html#present:Userna...)
2. Syntax and compiler errors are really confusing. When I walked around the Java class, most kids were asking questions about how to fix their syntax errors. In my class, most of the questions were conceptual.
It's easier to go back and learn a "real" language after you understand the high-level concepts and have seen how cool programming is.
If you want the resources I used, feel free to ping me ([email protected]).
If climate's your thing, you can check out https://www.climatechange.ai/. I'd bet similar communities exist around other areas for social good too. Good luck :)