I can see both sides of the argument, but I do think it is worth trying, and not just with one test of 30 people. If it could be tried out on different groups of people in different locations, of different ages, different family circumstances etc. over, say, the next five years, it could be very useful. It has long been an idea in circulation and just because it didn't seem successful in one trial doesn't mean it's not worth trying it again.
I think that more and more low paid jobs will be done by AI; it seems as inevitable as the effects of climate change! If we could find a really good way of running the scheme with the least loss to the poorer parts of society, it would be very worthwhile. It could improve the lives of the disabled, carers, fostered children leaving care, and the mental health of many.
I know a couple who have fallen through the net with an abusive former partner, child benefit being wrongly claimed by other family members, a child with a severe illness, which means frequent hospital visit with all the expense that entails. They work their socks off to keep going, and at the same time are kind and helpful to others. Having some money coming in would remove or at least reduce the daily stress and worry they have.