I'm skeptical that there's enough to prevent validators from acting maliciously here. Firstly, the only leverage against a validator is that if they act poorly

I'm skeptical that there's enough to prevent validators from acting maliciously here. Firstly, the only leverage against a validator is that if they act poorly, others will no longer accept their currency, so their personal UBI is possibly lost. Secondly, the validator doesn't necessarily _gain_ anything from continuing to act in a trustworthy manner. Maybe they require membership fees or something, but that's not intrinsic to the system. Taken together, all it requires is for a validator to insert _one_ fake account into the trusted pool to break even, and _two_ fake accounts to profit significantly. For example, they could create 50 fake accounts at the start of the experiment and let them accumulate fake wealth on the side, then operate as a validator for two years, then introduce these 50 fake accounts into the trusted pool and exchange 100-man-years of UBI for actual value. This appears to be more than they would receive over a lifetime of acting in a trustworthy manner, it's trivial to pull off, and if anything it's a _conservative_ attack - if a validator can get a userbase of 10000 people, 50 fake accounts doesn't sound easy for another person to detect.