That is probably what has to happen in the long run. Open source might be part of the solution, but if, for example, some company that sells widgets is going to replace 500 customer service reps with AI, even if it is open source AI, that's is still 500 less jobs around. No specific AI tech company might make a profit due to open source AI software, but those savings will still accrue to the widget making company.
Unless you go further and mean the state needs to take over all aspects of the economy (socialist/communism style). That doesn't work now, I think due to human nature and what motivates people, but in a world where human labor is in low demand, it may be inevitable.
But I can't even imagine the chaos in such a transition from the current capitalism model. There will have to be a period of mass poverty and social unrest in order to drive such a drastic change. This is just as likely to lead to widespread war as change to the system. Hopefully, it won't happen in my lifetime so I can skip the chaos.
Actual robots replacing human labor are still some ways off (i.e. not in my lifetime) but I can see any human who currently sits in front of a computer while doing their job being replaced by AI during my lifetime. That may not happen, but it will probably be technically feasible.
The AI being freely available doesn't allow a particular widget maker to use capital for that as a "moat."
Indeed, that doesn't bar the widget maker from having other capital based moats. There are incumbent(first mover) advantages like branding, network effects, and switching costs that capital can use to get to first.
But in someways a new upstart with leaner operating costs can counterposition their way to compete.
Yes. The period of extreme poverty is my biggest fear as well (like what happened at the end of the 1800s). I'm not resigned to it being an inevitability, however.
I don't think Communism (big C) has a proper way to deal with scale and the calculation problem of procurement, unless it tries state controlled rationing. The history of that hasn't been great.
The mixed model cannot be avoided in practice. At minimum, the state needs to properly craft the framework for fair markets.
I don't have all the answers, but I think it's important to fight on this front too. It's a front I am most familiar with.
But there are a lot of simple lies on both sides of the aisle that makes dealing with it harder.
There are complicated uncertain truths about technology.
Poor people are starting business at a rate that hasn't been seen for a long time. I think we need to lean into that phenomenon and help them be more than just simple gig workers. This area is one potential area to level the playing with AI. But it'll need to be a very different from large closed frontier models and more like the pre-fear-hype version of open use-cased focused models--but not all the way.
If you let the big AI companies strip-mine the commons and then put $200 to $300 a month pay walls for access, then the above mechanism of poor gig workers to become real CEOs with the aid of AI gets out of reach.
