I’d argue that in many cases, it is still an issue of money.

Are there some jobs technology can’t solve for yet? Yes. But if the company invested in research, they may be able to solve those problems.

Are there some jobs that technology can solve for, but the solution costs more than human labor right now? Also, definitely, yes. (Plenty of jobs are automated in the US but still done by humans in other countries.)

The above person’s point, as I read it, is that if people have UBI, they will be less motivated to do those jobs so supply of available labor goes down and the cost of acquiring that labor goes up. Then the company has a choice: invest more in people or in technology. At that point, scales may tip to technology.

To put imaginary numbers on it:

• If a technology will take $1T in R&D and $100M each to roll out to each of 1,200 warehouses, then the technology cost is $1.12T.
• If each warehouse employs 1,000 people at $50k a year, a year’s worth of wages for 1,200 warehouses is $60B.
• Human labor costs the company more than just wages though, usually anywhere from 1.25–1.4x more. Let’s go with 1.25 for $75B.
• So that would take ~15 years for the technology to exceed the cost of labor.
• But if people had more bargaining power thanks to UBI and wages double, they’d have to pay $120B in annual wages.
• If they also need to provide better benefits to attract employees, the multiplier would go up; let’s say 1.35x. Now labor cost per year is $162B.
• So now it’s less than 7 years for the labor costs to exceed the tech investment.

Companies aren’t great at long-term investments because they don’t show immediate shareholder value. Reduce the length of time for an investment to pay out, and they’re a lot more likely to do that.