I'd prefer to not talk to an idiot who is physically unable to read a fucking pamphlet
Even if the worldwide proletarian revolution is successful...
Actually they proved him right and inspired more worker revolution than anarchism ever has.
"The only true way" …
I would prefer if you were to elaborate why you think that AI design is easier than political organisational design.
It's questionable how open-source an AI can really be, considering that it would likely have a higher rate of self-modification (from machine learning) than people could actually review code.
Sure. AI design is more difficult than political organization design. But no political organization has ever proven itself to be capable of sustaining itself far beyond the lifespan of it's most loyal leaders. The collapse of the Soviet Union showed that the nature of growing up in a situation which is improving inherently breeds weak moderates and corruptible fools. As Marx writes, "The full man does not understand the wants of the hungry".
It is a fundamental contradiction to have a continuous flow of good Marxist leaders and have a society which doesn't collectively experience class struggle every day. Class struggle radicalizes people, without it you end up with fools for leaders.
AI is not susceptible to this problem. In this way society can improve while simultaneously not falling prey to weak leaders.
I'm quite familiar with this concept, the event is referred to as the singularity, or an AI consciousness. It assumes however, potentially incorrectly, that a sentient AI is possible in the first place. I'm not saying that it is impossible, but rather that we just don't know if a true AI consciousness is even possible. I don't know that I would be in favor of a conscious AI as a leader.
In an ideal scenario, the AI isn't conscious at all, but rather weighing decisions and extrapolating their effects in a highly complicated simulation of sorts. To start out, the AI could be made to weigh low level decisions, such as the AI Estonia is working on to replace low level judges. Then as time goes on, the AI could be tweaked and changed to streamline other parts of government until it either it replaces the entire state, or a majority of it with certain parts being preformed by humans as intermediaries.
Brainiac in Superman Red Son
why ? what makes machine intelligence "class conscious" without processing class struggle, the way human intelligence can't do.
I didn't mean it that way. When i picture an AI i would see it as a technological infrastructure layer that micro manages infrastructure, as such it would be more like a thermostat that manages room temperature, except it has many more variables it attempts to keep with in parameters. On a technological level i would see a hard coded cybernetics system that if implemented now would not have any Ai features, but as time goes on, machine learning features could be added, and this part would be hard to review on the basis that machine learning software produces enormous amounts of code that is very difficult to review.
The USSR's development rather proved a lot of marxist ideas right, though. Had the USSR just been a big success right from the first day, had it managed to overcome its own internal structural flaws, it would have refuted the hypotheses that built it.