Neither $3b nor $300b are realistically unfathomable to me. I find them easy to consider in terms of the projects I can build if I achieved each of these amounts.
As an example I’d have to allocate somewhere between $50m to $250m to get people to vote on a California proposition. I’d need to spend $1.5b to create a wing of a major hospital. I’d need between $100m and half a billion to create a new K-12 school in my city.
These are large sums of money and are currently out of my reach so if AI doesn’t destabilize everything my best bet is to take the same approach each of my ancestors did. Move my children one level up the wealth ladder and hopefully give them the values that help them prioritize these actions and the optimal way of getting there. I think that involves some amount of compounding and then some amount of spending.