Pretty much all of this rests on the assumption that recent changes in software hiring practices are primarily driven by AI, and that such changes will continue

Pretty much all of this rests on the assumption that recent changes in software hiring practices are primarily driven by AI, and that such changes will continue and spread across industries without foreseeable limit. The reality is that higher interest rates hit software particularly hard because less venture capital is being thrown at traditional software development. When money is tight, cutting new hires decreases to push off layoffs, and when layoffs happen experienced potential hires become cheap, displacing inexperienced entry level hires. No one is telling their boss "reduce my budget, the AI is so good I don't need these people anymore" they are getting told by their boss "find a way to make due with 3 fewer people." We should expect overtaxed workers to try and find ways to utilize AI to take up some of this slack, and higher ups may spin a tale of increased efficiency, but the fact is AI adoption is a symptom, not a cause. The hiring decrease and layoffs happened at plenty of places that have failed to adopt AI as well. Given that the current situation is unique to the circumstances, it does not hold that software portends the fate of all white collar work. That being said, we can certainly expect AI to improve, and attempts to be made to replicate and improve upon any genuine efficiency gains made in the present experiment. But the fact is that while AI may make certain tasks easier, that will lead to reorganization of the labor force more than disappearance. When mechanization of agriculture reduced the labor required to produce enough food to sustain people, people stopped being farmers. It was a major societal shift, and there were certainly issues, but we don't have 90% of our population made up of unemployed farmers who can't afford to buy food, nor even a large percentage of the population who wants to farm but is forced to work a much less desirable job. Comparative advantage will guarantee people are still doing something. There will always be tasks which would benefit from human input, and there will always be more such tasks. We may not currently place much value in these tasks, but by virtue of AI doing the other tasks, the relative value of fully automated tasks will decrease and the tasks which require human labor will become more highly valued. In a world where the best paid people are ditch diggers, and ditch diggers can afford yachts because yacht production is fully automated, who cares what the wage of the ditch digger actually is? Wealth concentration is a concern, but not because it will make it impossible for the vast majority to live a decent life. Instead the economic lives (and likely socio-political lives as well) of these two groups will simply diverge. This is extremely concerning from a standpoint of justice, but it's really orthogonal to AI. We've had such aristocracies many times before - they arise because of a failure of social institutions, not technology. We've been on the path towards them long before AI came along, and there is no compelling evidence that AI has accelerated the process. As far as economics is concerned though, your quality of life will continue to improve, even if some billionaire's improves faster.