Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: AI-town, run your own custom AI world SIM with JavaScript

github.com

Post reply on HN
Print view View original

81–90 of 121 posts

Re: Show HN: AI-town, run your own custom AI world SIM with JavaScript

#81

Post by pelorat »

liorben-david wrote: post #26

This is awesome! Game idea to build on top of this: Table top deception type game where each agent has the goal of convincing the real users that they are in fact also real users.(So each agent is trying to pass a turing test). Every AI agent uses RL to optimally prompt their personal LLM for how they should chat with the human players. eg should they try to frame a certain person, should they play it dumb, should th…

It would be pretty easy to beat. One sure way to tell an LLM apart from a human is to type something nonsensical and optionally repeat it multiple times. A human would inevitably answer with something along the lines of "what the hell are you on about dude???", which is something you'd never see from an LLM.

Re: Show HN: AI-town, run your own custom AI world SIM with JavaScript

#82

Post by xyzzy3000 »

jwmoz wrote: post #77

Irks me when people shoe-in the word "democratize".

The etymology of the word implies rule by a demos (demos / kratos), a demarcated ingroup of people such as the citizenry of a country. And yet, so many of these 'democratising' technical endeavours keep control of the project strictly behind lock and key. There's nothing democratic about it - they just want popularity.

Is 'popularise' an undesirable word now? I wonder why the euphemism treadmill has had to advance in this way.

Re: Show HN: AI-town, run your own custom AI world SIM with JavaScript

#84

Post by l33tman »

lumenwrites wrote: post #79
icoder wrote: post #75

Earlier quoted context omitted.

Haha yes the most commonly used argument is (afaik) that if we could create a 'complete' virtual world, we would, and projects like this and how enthusiastic we are about it suggest that that's indeed the case. (The argument then of course continues that if we would, our world's inhabitants would some day too, ad infinitum. Given such a stack, the chance that we're actually still in the top layer is very small.)

My counterargument to this is that people simulating our world would be simulating all the bad stuff in it, which would be cruel and unethical. I think if humans were simulating us, they'd simulate a more utopian world. And if curious-but-indifferent aliens/robots simulated us, they wouldn't be wasting compute on simulating all the boring and uninteresting things we have in our world. I just don't see why would an in…

Whoever is simulating us might be so far ahead of us that they simply don't see or care about what is bad and cruel and tragic. It's like if we would simulate an antstack or a beehive. We would think it interesting but wouldn't really care about the ethics of ant-brawls.

Re: Show HN: AI-town, run your own custom AI world SIM with JavaScript

#85

Post by e12e »

pelorat wrote: post #81
liorben-david wrote: post #26

This is awesome! Game idea to build on top of this: Table top deception type game where each agent has the goal of convincing the real users that they are in fact also real users.(So each agent is trying to pass a turing test). Every AI agent uses RL to optimally prompt their personal LLM for how they should chat with the human players. eg should they try to frame a certain person, should they play it dumb, should th…

It would be pretty easy to beat. One sure way to tell an LLM apart from a human is to type something nonsensical and optionally repeat it multiple times. A human would inevitably answer with something along the lines of "what the hell are you on about dude???", which is something you'd never see from an LLM.

> which is something you'd never see from an LLM.

Might not see it from ChatGPT - but "never from an LLM"? Why would you think that?

Re: Show HN: AI-town, run your own custom AI world SIM with JavaScript

#86

Post by pyinstallwoes »

fnordpiglet wrote: post #66
tavavex wrote: post #60

This is a very interesting use case of the Generative Agents project. I feel like this is just one of the earlier attempts at creating an interactive experience with this, but there's still so much potential for future games, if the technology is refined further. Think of something like Animal Crossing (or any other game where interacting with NPCs is at the forefront of gameplay), but with the characters that have c…

I feel like this might be the beginning of a much richer single player game genre that mixes generative language, image, texture, and other features with procedural rogue like world generation with the semantic features defined by the LLM. Obviously NPCs, dialogues, etc, can all be context aware agents bootstrapped by the environment and player history. Content creators IMO will be creating assets that are used to fi…

Which ultimately ends with thinking as a generative prompt into thoughts which get passed to the universal rendering engine known as reality.

Re: Show HN: AI-town, run your own custom AI world SIM with JavaScript

#87

Post by pyinstallwoes »

lumenwrites wrote: post #79
icoder wrote: post #75

Earlier quoted context omitted.

Haha yes the most commonly used argument is (afaik) that if we could create a 'complete' virtual world, we would, and projects like this and how enthusiastic we are about it suggest that that's indeed the case. (The argument then of course continues that if we would, our world's inhabitants would some day too, ad infinitum. Given such a stack, the chance that we're actually still in the top layer is very small.)

My counterargument to this is that people simulating our world would be simulating all the bad stuff in it, which would be cruel and unethical. I think if humans were simulating us, they'd simulate a more utopian world. And if curious-but-indifferent aliens/robots simulated us, they wouldn't be wasting compute on simulating all the boring and uninteresting things we have in our world. I just don't see why would an in…

My counter argument to that is try designing a more fair universe given all the known “natural laws” to exist. It’s pretty balanced from a “finely tuned perspective” to the extent everything exactly depends on the other.

Re: Show HN: AI-town, run your own custom AI world SIM with JavaScript

#88

Post by pyinstallwoes »

tetris11 wrote: post #78
icoder wrote: post #75

Earlier quoted context omitted.

Haha yes the most commonly used argument is (afaik) that if we could create a 'complete' virtual world, we would, and projects like this and how enthusiastic we are about it suggest that that's indeed the case. (The argument then of course continues that if we would, our world's inhabitants would some day too, ad infinitum. Given such a stack, the chance that we're actually still in the top layer is very small.)

There is something nice in the sense that simulation appears to be completely unmonitored. If my Sims started producing scans that approximated the hardware of my Gaming PC, I would probably stop the sim... assuming I was closely monitoring.

Why?

Re: Show HN: AI-town, run your own custom AI world SIM with JavaScript

#89

Post by pyinstallwoes »

PUSH_AX wrote: post #74
HeartStrings wrote: post #65

Could we not be living in a more advanced version of this same project?

This just reminded me of Sam Harris’s arguments regarding free will not existing and certain thoughts that seemingly originate from nowhere. I guess an LLM is injecting them!

The Bible begins with a declaration of recursion stating it’s a LLM.

In the beginning was the Word, and the Word was with God, and the Word was God.

Re: Show HN: AI-town, run your own custom AI world SIM with JavaScript

#90

Post by pyinstallwoes »

icoder wrote: post #76
Imnimo wrote: post #58

Here's a chat message from a random character in the demo I clicked on: ALEX 8/15/2023, 1:53:43 AM Absolutely! Here's a glimpse of my latest masterpiece. [Attaches a photo of the painting] What do you think? === I feel like it will be difficult in general to prompt the LLM in a way that gets it to stick to the limits of the simulation environment.

One solution could be to just extend those limits?

What’s a limit?
Post reply on HN