No, doing UBI via cryptocurrency is just a bad idea. To start with, UBI fundamentally needs to be centralized because a decentralized system can't determine who the unique humans are. You need to either rely on existing centralized government databases (which may also be flawed/corrupted...) or you need your own centralized testing personnel/devices.
What about something like
uPort? I'm not sure if the project is already operational and I'm also not quite sure if that could actually be used to tie you to a real-world identity.
https://medium.com/uport/the-basics-of-decentralized-identity-d1ff01f15df1Second, a decentralized cryptocurrency can only make money for "the system" through very blunt/stupid methods such as inflation and transaction fees, and this doesn't scale much. People won't use a cryptocurrency that's going to inflate a lot: that'd be stupid, and people aren't that stupid.
I hadn't even mentioned crypto UBI 'generating' money through inflation and transaction fees. Seemed pretty obvious that something like that wouldn't work, though then the question arises why there's a need for a crypto-based UBI in the first place, indeed.
The main point of my post was that the majority of these UBI crypto projects are almost fully reliant on either subsidization or grandiose plans to create a cryptocurrencies with significant standalone merit (like you mentioned). I would still like to hear some counterarguments from other people, but I'm afraid that there might not be many counterarguments to offer.
IMO a better idea would be to compete with Patreon: subsidize non-traditional work so that people can escape the 9-to-5 nonsense. But even then, I suspect that creating a new cryptocurrency would be counter-productive; you don't need to shoe-horn a new coin into every little thing.
Something like that already kinda exists on the Steem blockchain, but only for a specific subset of non-traditional work. There you have the
Utopian Project, which allows contributors to Open-Source projects to earn rewards for their voluntary work. I reckon that it would be possible to fork the project, to also apply to other sorts of work. Everything is manually moderated, so there's definitely the possibility of bias towards certain contributions/contributors.