So this may be true, but I think there's good reason to think UBI is more efficient than most aid programs for much the same reason free markets are usually more efficient than planned economies. Targeted systems tend to suffer from a few inefficiencies:
1. The bureaucracy consumes a large amount of money employing people and maintaining infrastructure
2. The burden placed on recipients of the money to ensure they meet the requirements lowers the efficiency of the aid. Even a few hours of appointments a month can end up being very costly in practical terms when you're poor.
3. The more the aid is constrained (e.g. a special card etc.) the higher the likelihood that it doesn't meet the need of the recipient as well as cash would.
The attentive reader may not all three of these inefficiencies are coming from the state's desire to more information about people's needs and wants. I can speak from firsthand experience that each of this issues can drop efficiency substantially.
We must also consider that many targeted programs are... not well considered, shall we say. We've known for decades that putting hard cliffs on benefit based on eligibility criteria is disastrous and yet it keeps happening. I have more than one disabled friends stuck in poverty wage they'll lose medicaid and instantly go bankrupt. This is wildly inefficient.
This is even assuming the eligibility criteria make sense in the first place, which is not a given. All that said, you could still be right, but I think there's enough possible upside it's still worth looking at very seriously.
UBI is not only more humane but also creates less overhead than the alternative of the government having to figure out not only how to allocate the funds, but how to actually distribute it to people that need it. Welfare is a complex system and there's a lot of available funding to different "targeted programs" but the burden of knowledge is placed on the people who need the assistance to seek those out. I find that burden unreasonable.
> There is of course some concern about how you'd implement a negative income tax
The machinery for this already exists due to the EITC. Expanding it to people without employment is a valid concern, but tax rebates based on income are not a totally foreign idea.
Giving a person money and then telling them to invest it wisely is not necessarily the worst idea ever.
The risky part is that you're giving the money to random people; as opposed to smart, promising entrepreneurs.
The question becomes what's the risk of free riding & unwise investment, versus indifferent investment, versus people who indeed invest wisely.
Personally I think that -in most of the OECD- the average/net risk already lies somewhat on the wise investment side. (slightly out on a limb: else how could the economy grow to begin with?)