The 1990s vision of computing was a bicycle - or car - for the mind. It was libertarian in the sense that if you had a device it would empower you to get where you wanted to go more quickly.
And the rhetoric around it was very much about personal exploration on a new and exciting frontier.
The 2020s vision is more like a totalitarian transport network where you don't own the vehicle, you don't own the network, there's constant propaganda telling you how to structure your journey to the standard destinations, and deviation is becoming increasingly impossible.
The device is just an access port to the network. It's dumbed down, so even if you understand how it works you can't do much with it. And as AI becomes more prevalent, your ability to understand that will diminish further.
So the end result is very plausibly a state where you're completely reliant on AI to do anything. And AI is owned by the pseudo-state oligopoly - the same oligopoly which runs the propaganda networks that sell you ads, hype selected content while suppressing other content, and genrally try to influence your behaviour.
It's the complete opposite of the original vision.
Will consumer AI fix this? Probably not. Even if the hardware keeps improving - debatable - a personal device is never going to be able to compete, in any sense, with an international network of data centres.
And this is where the geopolitical aspect comes in and where an increasing number of studies calls this 'Digital Authoritarianism' with the stated goal of a nation or company (or both in cooperation) keeping control of the population, the narrative and the access to information.
An overview of the literature and studies on the subject: https://www.tandfonline.com/doi/full/10.1080/02681102.2024.2...
A recent study that implicitely inverstigates the role of corporations in the trend: Digital Authoritarianism: from state control to algorithmic despotism https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5117399&... It's a bit long(ish), 29 pages (the last 10 are references) but worth a read.
The device becomes a magic artifact. Like a palantir. Many fantasy stories look like there were (or still are somewhere out there) great people who made all the magical stuff in the story while the people in the story have no idea how that stuff works.
That is possibly the way our civilization going. Especially when the datacenters will be in space, and only the "dumb" Starlink like terminals on Earth.
There's one way to deal with this, but I doubt it'll be popular in these parts: Communal ownership of the means of production.
Don't use the oligarchy's AI. Your personal hardware is going to be too weak. But together, we can own our own server farms.
The quote is a direct reference to a core tenet of Marxist theory, socialism, and communism.
Historically, communal ownership at scale has almost always been implemented via a centralized state, which has tended to gravitate towards authoritarianism. The Soviet Union and East Germany, and many other countries along those lines, didn't really fit the "hippy co-op" image very well.
If AI can code, and empower individuals to do it on a local device, it is already smart enough to educate masses on the matters of their self-interest, such as freedom and solidarity.
I don't think the powers will be able to gatekeep it. There might be some grief but overall human freedom will prevail.
They don’t know what they could have or why the new captcha is funny, thus they can never come up with a prompt that leads to them being educated on the matter. They would have to know that they don’t know and since there is no public discourse for such matters in their Facebook timelines, their thinly right wing digital news outlets and their Viber and what’s app chats they will never know that they don’t know.