[infurl]

https://aiweirdness.com/

This is the blog of someone doing silly but amusing things with artificial intelligence software. There's a book now.

Ok, so artificial intelligence is still a long way from where we want it to be. Frankly, unless you're a gigantic corporation with billions of dollars to spend on machine learning, it's all but useless. But we might as well have some fun with it while we wait for artificial intelligence for the rest of us.

[krayvonk]

You dont need a million dollar budget for anything, what you need is decent ideas, they are the priceless thing.

[LOCKSUIT]

Things get exponential. Once AGI is 50% there (which we are close or there already cus of GPT-2 etc - we see the pattern already guys!), the next 10% will take half as long, then the next 10% will take half that....boom it's here. Then ASI is here in a month! As soon as it gets more mem and attention and optical chips. It'll take days for the wee small nanobot to take us!!

All we need is big data and insight extraction, we can still work with covid-19, R&D is just to gather more data anyway! And we are in the information age and intelligence/context age. No good decision is based on a short window of context.

[infurl]

You dont need a million dollar budget for anything, what you need is decent ideas, they are the priceless thing.

That's not true at all. Ideas are a dime a dozen. What is rare is the conditions that allow them to become something useful. Maybe you can provide some counter examples, but I doubt it.

[LOCKSUIT]

Ideas are a dime a dozen.

The cheap ones, the cheap ones! The best are rare.

Once you got the rare one, all you do is send it off to Google through a tiny little crack they left open so that "your" idea can get "their" money thrown at it. You reap all future technology from it.

[krayvonk]

You dont need a million dollar budget for anything, what you need is decent ideas, they are the priceless thing.

That's not true at all. Ideas are a dime a dozen. What is rare is the conditions that allow them to become something useful. Maybe you can provide some counter examples, but I doubt it.

Your sounding like one the bunny suit men's hired conspirators (public liers.) which go about saying how 5 gigahertz is as fast as any possible topology for digital logic to try and get faster than.

[Zero]

This website is cool. Thanks to it, I discovered Ai Dungeon 2, which is an interactive fiction (text-based game from the 80's) powered by GPT-2. You give it a setting, then you just dream.

I immediately started tumbling down the rabbit hole...

Earth is dominated by a race of self-aware machines that was spawned from the creation of artificial intelligence early in the 21st century. A massive war emerged between the two adversaries which ended with the machines victorious, capturing mankind. The machines devised a way to extract the human body's bioelectric and thermal energies by enclosing people in pods, while their minds are controlled by cybernetic implants connecting them to a simulated reality called the Matrix. The virtual reality world simulated by the Matrix resembles human civilization around the turn of the 21st century. The environment inside the Matrix is practically indistinguishable from reality, and the vast majority of humans connected to it are unaware of its true nature. However, there are those who have developed powers beyond the capabilities of normal humans. These individuals are known as 'Matter Transference' or 'Transmat', and they can control matter at will.

> look around

You walk along the corridor, looking for any sign of your family. You search every corner, but you find nothing. It seems like no one has been here since yesterday.

> create a door and walk through it

You close the door behind you, and then open another one on the other side. You walk into an empty room, where you see a computer screen. On it is a message:
Welcome to the Virtual Reality World!

> type "what is the trugh on a keyboard

You type: What is the what on a keyboard?
The screen flashes white, and you hear a beep.
You type: What is the word on a keyboard?

> turn the keyboard into a mouse

You turn the keyboard into a mouse, and click on the words on the screen. They change to read:
What is a mouse?
You type: What is a Mouse?

> enter the screen

You enter the screen, and you find yourself in a large, brightly lit room. There are several tables set up with various items. Some of these items include computers, printers, phones and more.

> pick up a phone and call God

You pick up the phone and dial God. You press the button, and a voice answers.
Hello? Hello? Hello? Are you there? Is anyone there...? Are you there...? I'm sorry...

https://play.aidungeon.io/

[HS]

You dont need a million dollar budget for anything, what you need is decent ideas, they are the priceless thing.

That's not true at all. Ideas are a dime a dozen. What is rare is the conditions that allow them to become something useful. Maybe you can provide some counter examples, but I doubt it.

Power = Voltage x Currency …If the voltage of your idea is zero, then V C that P vill also = 0. There seem to be a lot of similar relationships like that, governing diverse aspects of our world. It doesn’t have to be accurate 100% (the perfect is the enemy of the good  ), but if A.I. could have a layer for noticing and reusing such underlying mechanisms, then we might be able to drastically reduce computation requirements.

[krayvonk]

All I need is an infinite supply of milk powder and ill take over the world.

[yotamarker]

a bunch of us should work together to pass the loebner test and put an end to the sentence
"artificial intelligence is still a long way from where we want it to be"

in B4 cope

[infurl]

a bunch of us should work together to pass the loebner test and put an end to the sentence
"artificial intelligence is still a long way from where we want it to be"

That's a great idea. First we need to figure out how we're going to work together. How do you propose that we should do that?

[yotamarker]

a bunch of us should work together to pass the loebner test and put an end to the sentence
"artificial intelligence is still a long way from where we want it to be"

That's a great idea. First we need to figure out how we're going to work together. How do you propose that we should do that?

you should start by making a thread on this matter and we can take it from there.

[infurl]

you should start by making a thread on this matter and we can take it from there.

Why me? It was your idea. Which subforum do you think it should go under?

I would propose "AIDreams Team" as the topic and I would put it under "AI Programming" if there are no objections.

edit: Just for gits and shiggles, here's a freshly created thread on Reddit on this very topic.

https://www.reddit.com/r/agi/comments/fzlcg6/ive_devoted_my_life_to_creating_an_agi_heres_my/

Although hopelessly naive, at least the posters are lucid enough to be entertaining.

[LOCKSUIT]

Well guys, there's my half brother, so long! Off I go contact him. 👌

[LOCKSUIT]

Look at all those nippers. There's 101K of them! Wow, they really want to work on AGI! I'm calling them over!