Skip Navigation

It’s Time to Stop Taking Sam Altman at His Word

www.theatlantic.com It’s Time to Stop Taking Sam Altman at His Word

Understand AI for what it is, not what it might become.

It’s Time to Stop Taking Sam Altman at His Word
103

You're viewing a single thread.

103 comments
  • but for now, his approach is textbook Silicon Valley mythmaking

    The difference is that in this case it is not hype—it is reality. It’s not a myth, it is happening right now. We are chugging inevitably down the track to the most dramatic discovery in human history. And Altman’s views on solving the climate crisis, disease, nuclear fusion… they are all within reach. If anything we need to increase our speed to get us there ASAP.

    • Tell me honestly, are you a bot or do you sincerely believe this shit and based on which qualification and experience?

      Gunpowder, electricity, combustion engines, universal electronic computers, rocketry, lasers, plastics - none of these made any dramatic changes. It was all slow iterative process of fuzzy transitions and evolution.

      While these made pretty fundamental impacts. Sam Altman's company is using fuckloads of data to calculate some predictive coefficients, and the rest of its product can be done by students.

      It's just real-life power controllers trying their muscles at bending the tech industry with usual means - capturing resources and using them to assert control. There were no such resources in the beginning, and then datasets turned into something like oil.

      Generally in computing (when a computer is a universal machine) everyone able to program can do a lot of things. This makes the equality there kinda inconvenient for real life bosses who can call airstrikes and deal in oil tankers.

      There was the smart and slow way of killing that via slow oligopolization, but everyone can see how that doesn't work well. Some people slowly move to better things, and some were fine with TV telling them how to live, they don't even need Internet. All these technologies are still kinda modular and even transparent. And despite what many people think, both idealistic left and idealistic right build technologies for the same ultimate goal, so Fediverse is good and Nostr is good and everything that functions is good.

      So - that works, but human societies are actually developing some kind of immunity to centralized bot-poisoned platforms.

      To keep the stability of today's elites (I'd say these are by now pretty international), you need something qualitatively different. A machine that is almost universal in solving tasks, but doesn't give the user transparency. That's their "AI". And those enormous datasets and computing power are the biggest advantage of that kind of people over us. So they are using that advantage. That's the kind of solution that they can do and we can't.

      Simultaneously to that there's a lot of AI hype being raised to try and replace normal computing with something reliant on those centralized supply chains. Hardware production was more distributed before the last couple of decades. Now there are a few well-controllable centers. They simply want to do the same with consumer software. Because if the consumers don't need something, they won't have that something when they see a need.

      All these aside, today's kinds of mass surveillance can't be done with (EDIT:without) something like that "AI". There simply won't be enough people to have sufficient control.

      So - there are a few notable traits of this approach converging on the same interest.

      It's basically a project to conserve elites. The new generation of thieves and bureaucrats wants to become the new aristocracy.

You've viewed 103 comments.