This is a diversion from the main thread, but that's fine. I want to restate that I am rather worried about Automation-Induced Unemployment (AIU) running away on us, and I see a scaled UBI as part of the solution (along with targeted employment for positive externalities).
You will always need additional welfare. For a large number of poor people, the main thing they need is just more money. They have all the life skills required to make a better life for themselves with a larger paycheque. And there's a smaller segment of poorer people that both need much larger amounts of money and actually targeted assistance. There's just no way to balance that spending with the amount of money everyone else receives. The major problem with much of Western welfare isn't that welfare is given, it's the perverse incentives due to their clawback mechanisms.
The positive reason for UBI isn't solely its ability to create efficiencies in our welfare systems (though it very much would, by large margins), but also to counteract the systemic imbalance we have in our societies with too much wealth having gone to the top. If it wasn't for AIU, an extended period of (high enough) UBI would claw back our societies on track. You don't need to eliminate welfare programs for a UBI system to be of use.
You haven't provided any numbers, you realize. There's only so much insistance that you're correct that's really useful to provide in a discussion. Also, I know it was an interruption of the thread, but what's under discussion is whether we can afford a UBI where people can choose to not work. Inflation isn't really part of that story, except insofar you need some mechanism to make sure everyone has the same (minimum) purchasing power today as next year. But that's easily modeled, even if not implemented.