The Core

People generally have no idea that power can jump to $3/KWH (aka $3000/MWH).

The spot prices on the wholesale market fluctuate so much. If this were passed on, people would be installing batteries like mad, which would eventually reduce the problem and thus the value of the batteries. But realistically, I think among the upper 25%, major battery banks are going to become normal in new home builds, especially custom homes.

As the Romans said: private luxury, public squalor. That seems to be working its way through every system in the US.

“AWS accounted for 62% of total operating profit”

!!!!

Amazon is basically a web services company plus an ad marketplace with a retail operation on the side.

I’ve been thinking about a related problem - the inherent unhappiness of people who try too hard to optimize in life (I generally refuse to use the phrase over optimize because logically that is impossible).

Here’s Freddie’s take on the social downsides of optimizing services.

My current working definition of overoptimization goes like this: overoptimization has occurred when the introduction of immense amounts of information into a human system produces conditions that allow for some players within that system to maximize their comparative advantage, without overtly breaking the rules, in a way that (intentional or not) creates meaningful negative social consequences. I want to argue that many human systems in the 2020s have become overoptimized in this way, and that the social ramifications are often bad.

https://freddiedeboer.substack.com/p/the-modern-curse-of-overoptimization?utm_source=multiple-personal-recommendations-email&utm_medium=email&triedRedirect=true

How do you know that?

I mean, how do you know which pages are PHP pages?

Exception: politics.

I also like Derek Sivers idea of Version Infinity. Instead of envisioning Version 0.1 which would be enough to test the concept, people often end up trying to launch Version Infinity, the one with so many features it will just knock everyone’s socks off. If only it could ever be specced, built and shipped

True, though I do find us remarkable in what we can accomplish

For sure. We are clearly radically different from any other life form on the planet. It boggles my mind that we can send spacecraft to other planets, discover elementary particles, develop coherent ideas about the age of the universe and invent Cheetos and Doritos.

But the ancient designations of man is the "rational animal" or "homo ludens" (the human who plays), "man the toolmaker," and so forth seem hopelessly outdated. We found out that these things that defined the "essence" of humans are either not unique to humans (tools) or don't describe humans particularly well (rational).

What is intelligence and what is life?  When an AI has self-awareness is it alive?

Also spot on I think.

My guess is that there will be an effort to define those things in a way that protects human exceptionalism until it is simply no longer tenable.

Somehow, we need to protect the idea that a non-human animal that uses medicinal herbs for healing and a non-human computer system that writes beautiful novels and provides an excellent course of treatment for a sick patient and effective, compassionate counseling for people with mental illness nevertheless have essentially no rights, whereas the most despicable humans nevertheless do.

It will be an interesting debate and in the long run I wonder whether it will be hard to coherently separate the debate on the rights of non-human synthetic life from the debate on the rights non-human biological life.

But frankly, the future has a way of surprising and predictions here are especially going into uncharted territory. Curious to hear more thoughts though

I was just thinking about things like this in the context of AGI.

We don’t have “real” AI because every time we solve a classic AI problem, someone says the solution is not real intelligence. But we’re getting to the end of that line

And that reminded me of how we used to redefine human exceptionalism every time an animal was shown to do a “human” thing like use tools, count, recognize themselves in a mirror, lie.  Now things like another primate using médecine is only so surprising, but it’s not like Jane Goodall discovering tool use and murder.

Within the biological world, human exceptionalism is mostly over. There doesn’t seem to be a uniquely human characteristic.

I see the same process happening in the AI world.

Attention: You are now leaving the Uncanny Valley (but you are not quite out)

I’d love to be part of a double blind test to see if that feeling is because I know it’s generated. I feel that if I watched an hour of that I would feel anxious or tired and not know why because I wouldn’t know it was generated, but somewhere I would feel it.

I am guessing that a year from now at most, the best models will be fully out of the Uncanny Valley