This is something I have been thinking about for quite a while. I readily embrace the advances AI may bring to mathematics and the hard sciences, but those advances were largely expected even predictable.

What I had hoped for was something more ambitious: using AI as an arbiter in economic, political, and social debates, one capable of weighing evidence, exposing trade-offs, and helping us make decisions that produce better outcomes over the medium and long term, even when those decisions conflict with powerful private interests.

I suspect, however, that this is not a particularly urgent goal for the people funding and directing these systems, many of whom live far removed from scarcity and its consequences.

Not really. They're an AI company: they develop AI and sell it. There's some room for pulling off flashy marketing stunts, but not all that much.

This is division of labor, and it's a good thing. I'm sure OpenAI employees, who are very well paid, donate some money from their salaries to others working on the areas you're citing: probably more than you think, I say that from having attended some EA parties back in the day.

But that isn't my point: my point is that a company which makes brushless motors should put most of its time and money into solving the "make and sell brushless motors" problem, and if they or their investors feel like they need to do more for the world, give money to the people who have the time and ability to do things about that. There are a lot of quality-of-life improvements which need brushless motors.

Next question is how useful their product (OpenAI, I mean) is to more focused do-good-in-the-world professionals. I'm sure that varies quite a bit. For getting the homeless off the street? I conjecture, not very useful. For 'complete the transition off fossil fuels'? Extremely useful, no one in that field knows how to do their job without AI in summer 2026. I'm certain of this.

Solving math problems doesn’t require the cooperation of rival factions.

None of these problems need technology to solve. Step 1 is trying to get different groups of people to cooperate with each other. Good luck with that.

Yes, but all of those are orders of magnitude harder than math.

What's stopping the non-profits that exist today from just putting more money into tokens to get the solutions they want?

Seriously?

We already know how to solve all of these issues. What we lack is collective political will.

Almost like superintelligence solves these problems...

We know how to solve food insecurity (in 1st world countries). We have plenty of food. Capitalism requires though throwing out food that can't be sold because billionaires find giving away things anathemic to their worldview. Get rid of billionaire sociopaths.