The NIT does seem like a reasonable alternative to TANF and other welfare programs that have lost effectiveness in fighting poverty. From my research, (and take this with a grain of salt,) the EITC seems to be able to alleviate poverty while inspiring work in a very similar fashion to the NIT. It functions as such:
Enrollees start having a portion of their taxes covered the moment they earn their first dollar. Coverage increases steadily for a while as the enrollee accumulates more earnings. Eventually, these enrollees pass a certain threshold at which point, coverage is phased out. At the end of the process, the recipient acquired a decent job and, (hopefully,) the job is sustainable enough so that future coverage through other transfer payments is unnecessary.
The National Bureau of Economic Research published a study in May of 2015 which analyzed existing research on the subject. They summarized the findings of multiple studies and consistently found an immense poverty-alleviating effect due to the EITC. This included data from the Census Bureau which projected that upwards of 9 million people, roughly half of whom are children, were lifted out of poverty due to the EITC (1). The EITC also targeted households within 75-150% of the FPL which helped target the most vulnerable families (1).
The system is not without its flaws, of course. Single people with no children do not reap as much benefit, with less coverage overall. While this makes sense from a policy perspective, as people with less children do not need additional transfer payments to accommodate them, the percentage paid is dismally low regardless. The maximum credit that a childless worker can expect to receive from the EITC is a measly $519 dollars per the Tax Policy Center (2). Additionally, payments in error plague the tax credit. A staggering 23% of EITC payments are in error, mostly due to misunderstandings as to whether a child qualifies under the EITC guidelines (2). Perhaps clearer instructions given to those who signed up for the EITC and an expansion of coverage for childless workers would be able to aid lower class people while incentivizing work.
Since you were worried about unemployment in regards to the NIT, perhaps this is a middle ground that can be reached. Tax credits would only be given to those who worked, and the expansion of such a credit could lift more people out of poverty. But of course, automation is going to kill more jobs in the short term, presenting an issue in which plenty of people are going to be looking for a job and are probably not going to be able to find one. Perhaps an addendum to the credit would be a plausible solution. As long as you can prove that you are taking steps to find employment through vocational training, education in a valid institution (as in, trade schools, universities, colleges etc.) credits could be distributed at a certain, fixed percentage until one finds a job.
Nevertheless, an NIT contingent on work might be the best solution if resources are available to aid people in finding employment. Of course, that is a big "if" and would require the collective efforts of political institutions to start doling out more student aid, improve the quality of education through more vocationally-driven classes, and create jobs through public works.
Additional steps that could be taken include actually hiring competent people to deal with children with physical, psychological, and emotional disabilities. Too often, people with severe disabilities are placed in segregated workplaces where they perform menial tasks under the pretense of "job building skills." They are often paid a sub-minimum wage, which is legal under 14(c) of the Fair Labor Standards Act (3). Pay could be as low as 40 cents an hour, and is often not heavily regulated due to a lack of investigators at the Wage and Hour division of the DOL (3) (4). Early intervention and social support could remedy many of the problems keeping disabled students back from work. Unfortunately, special education teacher employment has dropped rapidly, by over 17% in the last decade according to Education Week (5). Without enough teachers, the pool of possible candidates have shrunk for these positions. Thus, less competent teachers can receive employment in the ESE department out of virtue of not having to compete against talented teachers also vying for the position. While national averages of teachers considered "highly qualified" in special education has been relatively high, this measure tends to vary considerably by state. In Kansas, for instance, only 70% of special education teachers are considered "highly qualified" (5).
