Books with intelligent ai or transhumanism

Good to hear. Stapledon is an author I've never heard of so I'm excited to probe uncharted territory!

RIP humanity

Post a thread about it when you are done, I need more people to talk about it with.

LAFM is one of the most significant books I ever read. It fucking blasted my mind with all these possibilities and possible conceptions of thinking about humanity in the extremely long term, humanity in the loosest sense of the term. It just became a part of my mental machinery somehow.

It's also a great read. I can never bring myself to believe that it was so early in scifi's history.

is this good?

It's pretty good, it's fun if you're super autistic because he goes into gratuitously complex speculations about how to trickydick an AI into not fucking your shit up, but it has long stretches of minutiae too

Also he almost completely fails to address "mind crime" against AIs, rather than just against simulated human minds, and he fails to address certain cutting edge issues with AI research (like Dreyfus' phenomenological critiques)

Also his other views range from kinda-interesting "reformulations" (polite way to say it) of older scifi/thinkers to full retarded simulation theory stuff but that's just my taste

I listened to it as an audiobook while shelving in a library, and it was nice to zone out on.

Forgot to say though, at base it's simply an IMPORTANT book to read because it's a strong dissenting voice in the AI issue. As he documents, most AI people are autistic tinkerers who have very naive assumptions about what would happen if they succeeded at creating full-fledged AI. Bostrom has the balls to come right out with the fire and brimstone. It is the most terrifying existential problem this species can face, and we are on an uncertain precipice where we could stumble into it at any second (picosecond, more likely)

He deserves massive credit for that

>to full retarded simulation theory stuff
does nicky think we're in the matrix?

thanks for the informative posts

Blindsight by Peter Watts. One of the best sci fi books I have ever read.

I believe he thinks one of the following must be true:
1. We're in the matrix
2. No civilization ever becomes advanced enough to build the matrix (in the history of the universe)
3. No civilization who becomes advanced enough to build the matrix ever wants to build it (in the history of the universe)

Mah nigga. Got echopraxia waiting for me on my bookshelf. Blindsight really left an impression on me too.

The Sprawl trilogy.

sounds pretty reasonable.

The Culture series

Dune.

First two books of Hyperion are great, and the first one is GOAT. Can't recommend it enough.

Perfect Imperfection Jacek Dukaj,
Quite intriguing with fresh concepts.

So Veeky Forums?

Is strong A.I possible?

As a roboticist for hire, I believe and hope we'll see artificial general intelligence within this century. It's also my dream to partake in its creation.

the chinese room argument always triggered me because what it really does is cast doubt on the human capacity for understanding. Does Searle think humans have a little homunculus that does the "real" understanding?

Also like, the book would have to contain an amount of information comparable to a human brain. If it is assumed to be a regular book printed on paper, it would be impossibly huge.

I don't know man. I think the point is that even if the rulebook was shifted into searle's brain, and there was no room, searle still couldn't be said to speak Chinese, because receiving inputs, processing them, and producing outputs, doesn't account for human conciousness

I disagree. If the machine was indistuingishable from a human in conversation, which is a premise of the argument, it would have an understanding of its own responses. Whether the instruction set is followed by a (conscious) human or a computer doesn't matter, and whether the human understands what they're doing doesn't matter - the apparatus as a whole has understanding. It could not converse if it didn't.

>intelligent ai
If so will we RIP in peace?

You're in terra incognita here afaik. I would say tho that if you were to inspect some of the modalities of what you're saying you'd find you'd have to work out the relationship between machine/agent, language and understanding.

I believe that perfectly conversing in Chinese is an "AI-complete" problem so that any algorithm that manages to do it is an artificial general intelligence and as such would have an understanding of what it's speaking. How the algorithm is implemented is irrelevant as it's already assumed that it works.

Only drunk people can hold conversations without understanding what they're speaking and I'm fairly sure such conversations aren't what the Chinese room argument posits.

A Fire Upon the Deep. Very good concept.

Surely, the argument does show that a structure can be created that can converse without understanding. Unless we rely on a very esoteric understanding of the word understanding

Apologies if I've misunderstood what you're saying

No, I simply think that an appearance of understanding is no different from actual understanding. If the machine is capable of conversation, it must at the very least appear to have an understanding (otherwise it wouldn't be even nearly human-like). And if it looks like a duck and quacks like a duck, it's a duck. There's no mysterious component which enables humans to have "true understanding", we simply appear to understand things.

But there are two different kinds of understanding involved. Searle understands what is in the rule-book and how to connect an input to an output, but this is not what it is to understand a language. Our understanding of language involves an internal reflection on the sense of the symbols, whereas in Searle's case he only understands the input with regard to connecting it to an output.

Searle's own understanding is irrelevant, he's simply following the instructions. Replacing him with a computer doesn't matter. The instructions themselves enable the understanding. The instructions encode some algorithm (that in its entirety is almost equivalent to the full human brain, as conversation is an incredibly complex process) and that algorithm understands the conversation.

Will do, friend!

But if the system exists entirely in Searle's head then his understanding is indeed important. We're trying to determine if the appearance of his understanding the langauge actually reflects his understanding the langauge. I'd claim it simply represents his understanding the system of connections not the symbols themselves. To my mind, it's a sound argument

It's absurd to ask if the physical laws that govern the chemical reactions in your brain understand conversation, and whether that question has any bearing on your understanding.

Searle in the thought experiment only makes the instructions happen; he's a physical law. It's his rigorous following of the instructions that allow the algorithm to process information and understand. The only thing that matters is that the instructions are followed - what Searle thinks of the instructions or the Chinese conversation have no bearing on the algorithm's understanding.

But are you then saying something akin to language=consciousness?

Yes, language use is deliberate and conscious. Certain applications of language can be convincingly faked without consciousness - like producing poems - but open-ended discussion with humans is not one of them.

>Yes, language use is deliberate and conscious.
Not in the Chinese room.

Anyhow there's a nice youtube lecture with Honderich and Chomsky on language and consciousness.

>All this talk about language
>The arrival comes out in a few days

So is Story of your life good? And where can I read it? It's super expensive on amazon for some reason. (I assume limited printing+movie hype=low stock)

read Neuromancer, how is this thread not just composed "read Neuromancer"?

your taste fucking sucks, you have no idea what you're talking about, this is the most Veeky Forums post I've ever seen in my life