Ocasio-Cortez and Sanders introduce AI Data Center Moratorium Act

I’m not necessarily in favor of data centers. That said, every time I see something like this I just wonder when we are going to start hearing about data centers being built in poor countries powered exclusively by coal and zero regulation. I really have no idea what a good outcome looks like but playing ultra hardball seems unwise. I’m very poorly educated on data centers and AI in general. But, I am an expert in electrical generation. I know we could do it pretty clean, at least relative to coal/heavy oil. I don’t think stopping data centers built in the US will magically cease the boulder rolling in the AI direction. Seems like the money is at its back so its going to happen, just a matter of when, where and how dirty.

You basically understand what the people with a vested interest in making AI happen want you to know. The truth is that AI is already starting to crumble. It’s a technology that doesn’t do 99% of the things it’s perported to do, and will never do 90% of what they sold it on.

Yeah I’m not versed in the subject enough to say/think you’re wrong necessarily. I do know the general slant Lemmy’s population has against it though.

GenAI as it currently stands is a fancy text predictor. You ever had your phone suggest the next word in a message you’re typing? It’s that, on crack.

When you really wrap your head around the fact that that is all it’s doing, it loses a lot of its appeal imho. Especially for the cost to do so.

To be more specific (for anyone interested), the next word predictors are usually a type of model called an LSTM (at least I think that’s the most common). This model type has been used for a long time for dealing with sequential data. In 2014 there was a famous paper introducing an attention mechanism. This was a rather brilliant, though relatively minor extension to how LSTMs work. Essentially between each step of an LSTM it generates some data representing the model’s knowledge of the sequence to that point. The attention mechanism looks back at these intermediate values and determines how relevant each state is to the current point in the sequence and pulls in the most relevant bits. This vastly improved the memory of the LSTM over longer sequences.

In 2017 there was another famous paper “attention is all you need” which said something to the effect of “the attention mechanism is doing all the work, we don’t need the rest of the LSTM we can replace it by running attention between all point combinations in the sequence.” It’s actually significantly slower to run as the model grows, but much much faster to train because it’s not intrinsically sequential. This is the transformer model that’s the basis of all our LLMs.

Obviously some massive simplifications here but as despite being fairly anti AI, I do love the engineering behind it. So yeah, pretty literally a fancy text predictor, but it turns out when you throw all the compute you can muster at a fancy word predictor is makes the world go crazy