The term ‘Language Learning Model’ is as practically meaningless as it is technically accurate.

https://en.wikipedia.org/wiki/AI_effect

Pop in the elevenlabs and do a video again and ask for a patreon/donations

Picks up lines that fulfill the requirements

I am not saying that isnt impressive or useful, just that it isnt original really.

Yes, I actually think this is a good idea.

It gets very woo-woo very fast, and the focus needs to remain solely on science / building an actual product, but when studying cognition so unconvential approaches really help. In particular, altered states of consciousness help understanding the specifics of your brain processes. From here, two mains camps: entering altered states with psychoactive substances or without.

I personally fall in camp 1: psychoactives tend to impact various parts of the brain differently, giving you a ventage point to understand the different functional components of your brain, how they interact and what purpose they serve (cf. The thousand brain theory).

I have heard that folks achieve altered state through mediating / breathing / visualiazing, but it's hard to find people that also have the technical baggage to transform the insights they get into technical elements of an Architecture for AGI. If you know people who might, I'm all ears, tell them to read this :)

It's definitely on my radar. I wanted to start off with xVASynth first as the voice models are readily available / it's free, but I'm hoping to add functionality to switch out different language / voice synthesis models as needed to find the best balance between performance and cost.

Are you saying vegans don't get cancer?

Yeah, watching society anthropomorphize AI or, in some cases elevate to it to mythical status, as in deities is mostly endearing, who am I am I to deny someone uhhh putting googly eyes on thier toaster and considering it part of their family or the leader of their weird cult. Just make sure that sophisticated toaster of yours doesn't accidentally, or intentionally, ruin everything, and we may all be perfectly okay!

Yeah seriously, some of these responses had me questioning if it's me who is the NPC.

John von Neumann

>John von Neumann ( von NOY-mən; Hungarian: Neumann János Lajos [ˈnɒjmɒn ˈjaːnoʃ ˈlɒjoʃ]; December 28, 1903 – February 8, 1957) was a Hungarian-American mathematician, physicist, computer scientist, engineer and polymath. He was regarded as having perhaps the widest coverage of any mathematician of his time and was said to have been "the last representative of the great mathematicians who were equally at home in both pure and applied mathematics". He integrated pure and applied sciences.

^([ )^(F.A.Q)^( | )^(Opt Out)^( | )^(Opt Out Of Subreddit)^( | )^(GitHub)^( ] Downvote to remove | v1.5)

>(imagine thinking that no one understands what “intelligence” is except you😂), and speculative philosophal nonsense. (With a hint of narcissism thrown as well.)

I really get the sense lot of the time reading the doubters is that they think nobody else even considering all the problems and challenges. All these PhD researchers are just mindless idiots chasing some fad. Just reeks of an immense hubris.

>The author made the laughable claim that superhuman AI was merely science fiction

The thing is, this thing doesn't even need to be superhuman. I am not sure how many people know of John Von Neumann but he should have been as famous as Einstein and he was arguably even smarter. His Wikipedia page reads like piece of fiction, you look at his huge list of things he is known for and at the end of the list you have (+93 more), what... His contribution to mathematics and computer science is beyond immense, it's very likely we would have been right now quite a bit behind in number of fields if he didn't exist. Now imagine if instead of person of this kind of brilliance being born once or twice a century, we could instead have million of them, on-demand at all times. If it wouldn't result in a singularity, it would be something very close to it.

I only take issue with people trying to build a stable for their car instead of a garage because they feel bad for it.
And are trying to berate me for not acknoledging the feelings the car might have for being put in a cold garage.

Although I must admit that the aesthetics of a feathered plane might look bitch'n, I refuse to bring birdseed with me on a flight.
Because it's a machine, it's a tool. It's a pattern juggler of words of the highest degree.
But it expresses found commonalities it has been fed.
It's a mirror and a sifter of zettabytes of stories and facts.

But there is a loved narrative here that these tools are persons, and they are expressed by people who use the chatGPT in a way that steers the tool to this prevered conclusion.
And this is easy, because of all the data and stories that have been fed into the tool, stories about AI being persons are part of it.
So It will generate the text that fits this query.
Because we told it how to.

Yes and I think it will be even better when these NPCs will be able to "understand" the world around them : other NPCs and locations.

"It's just predicting the next word."

Nobody is claiming that AGI isn't possible. What people are skeptical of is the endless corporate PR that "We have created AGI" or "AGI is near". There are so many gullible fools believing in corporate PR of AI hype. It's beyond pathetic.

We will keep moving the goalposts until we have created God, then we get bored and start over again.

This is something I have burned so many calories thinking about and I somehow always end up in a place like this, typing vague ideas out to strangers on the internet, no better off than before.

Looking forward to all of the new AI generated Fleshlight designs though, going to be exciting regardless of all the existential dread I'll be pumping into them every waking second.

You're talking about how we don't need AGI in a Singularity sub? Jesus Fucking Christ, an AGI is the entire point of a singularity.

AI hypers: AI will be so smart that there will be either utopia or dystopia

Also AI hypers: Durr, AI can't be as smart as Einstein, that's moving the goalpost!

It's not that they cannot. It's that we still have no idea how sentience works.

And if you want to use a computer for your research, you guessed it bud, time to build a fabrication facility and re-invent the microprocessor.

Oh, you need the internet? You guessed it, ARPA 2.0 done by yourself.

Same. He he

The point is that it neither flies nor sails. It's basically "cargo cult science" where it only looks like a plane.

>LLMs are capable of completing functions that were previously only solvable by human intellects

That's only because they were already solved by the human intellect. It's only a mimicking machine.

Then where does the poetry come from?

I mean it sort of has, all the protein chain folding stuff was practically discovered overnight.

Gpt and every machine learning software learns messages, and then repeats them.

It doesnt learn words and concepts and use them autonomously for original thought.