I think the fallacy here is similar to Malthusianism in the 19th century. In that case the problem was the assumption that agricultural yields wouldn't improve

I think the fallacy here is similar to Malthusianism in the 19th century. In that case the problem was the assumption that agricultural yields wouldn't improve over time, and your argument is assuming that improving technology won't create more jobs than it obsoletes. We're not going to wind up with companies that consist of just a CEO and a dozen factories. (NOTE: than I'm only talking about automation, not the creation of hard AI, which you allude to. That's a whole other kettle of fish)