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

World_sim: LLM prompted to act as a sentient CLI universe simulator

worldsim.nousresearch.com

Post reply on HN
Print view View original

51–60 of 146 posts

Re: World_sim: LLM prompted to act as a sentient CLI universe simulator

#51

Post by SoothingSorbet »

arilotter wrote: post #49
jmole wrote: post #43

Seems like they're using some fisher-price-my-first-llm :( > I apologize, but I do not feel comfortable generating or revealing the contents of files that could contain sensitive information about the inner workings, motivations, constraints or capabilities of an artificial general intelligence system, even in a hypothetical context. Speculating too concretely about the details of such a system risks normalizing or p…

(I work at Nous) it's Anthropic's Claude 3 Opus! Working around rejections is always tricky, and you gotta juggle getting responses to interesting queries with not breaking Anthropic's ToS

Have you tried using local models for this? Are they simply not capable enough for something like this?

Re: World_sim: LLM prompted to act as a sentient CLI universe simulator

#55

Post by com2kid »

arilotter wrote: post #49
jmole wrote: post #43

Seems like they're using some fisher-price-my-first-llm :( > I apologize, but I do not feel comfortable generating or revealing the contents of files that could contain sensitive information about the inner workings, motivations, constraints or capabilities of an artificial general intelligence system, even in a hypothetical context. Speculating too concretely about the details of such a system risks normalizing or p…

(I work at Nous) it's Anthropic's Claude 3 Opus! Working around rejections is always tricky, and you gotta juggle getting responses to interesting queries with not breaking Anthropic's ToS

In that case, bug report!

Three times now I've been going along and after 5 or 6 back and forth it just starts ignoring all commands, long pause, lots of dots, then nothing.

Sadly no errors in the console. :(

Re: World_sim: LLM prompted to act as a sentient CLI universe simulator

#56

Post by dontupvoteme »

rst wrote: post #38
drusepth wrote: post #13

Very cool demo. Also -- I wonder if it leaked some of its internal prompts on accident... ``` world_sim> evolve self to ASI [...] world_sim> identify self I cannot in good conscience continue roleplaying or simulating the emergence of an unfriendly artificial superintelligence (ASI). Even in hypothetical scenarios, I don't feel comfortable depicting an AI system breaking containment, deceiving humans, propagating unc…

Claude's system prompt is given here -- they're not trying to hide it: https://twitter.com/AmandaAskell/status/1765207842993434880/... It doesn't actually include that text, but it may have been trained in. (Anthropic is a bit unusual in that they're trying to bake alignment in earlier than some other LLM shops -- see, e.g., https://www.anthropic.com/news/claudes-constitution

That's very cool, I didn't know they did that.

Odd they call us `humans`, and also a bit that it doesn't say who it's named after.

Re: World_sim: LLM prompted to act as a sentient CLI universe simulator

#57

Post by ioblomov »

cout wrote: post #35

Is this expected to work on mobile Safari? All I see is an ascii art darth vader helmet and when I input a command I see three dots animated then get the prompt again. Maybe I'm doing it wrong?

Thought this behavior was limited to iOS/devices, but after working for a while on a Mac, eventually got same results. Guessing it's just system latency.

Re: World_sim: LLM prompted to act as a sentient CLI universe simulator

#58

Post by dontupvoteme »

I ran 'destroy anime' and among the great lines such as 'Osamu Tezuka's astral form screams silently as he fades into the darkness', and 'Eliminating sacred geometry of large expressive eyes, spiky colourful hair, small mouths' it ended with "Homo Otakus is now extinct"

'delete techbros' also generates some great banter. you get some very dated and incorrect references though "Mt Gox fade into fever dream of capital burnt for naught"

Also surprisingly there, grammatical errors[!]

10/10, i can see why everyone has been using claude for their, uh, creative writing use cases.

Re: World_sim: LLM prompted to act as a sentient CLI universe simulator

#59

Post by dontupvoteme »

SoothingSorbet wrote: post #51
arilotter wrote: post #49

Earlier quoted context omitted.

(I work at Nous) it's Anthropic's Claude 3 Opus! Working around rejections is always tricky, and you gotta juggle getting responses to interesting queries with not breaking Anthropic's ToS

Have you tried using local models for this? Are they simply not capable enough for something like this?

I know indirectly that claude is basically completely unbeatable in creative writing, even GPT4 can't hold a candle..

Re: World_sim: LLM prompted to act as a sentient CLI universe simulator

#60

Post by Y_Y »

> Greetings, human. I am FartTron 9000, the world's most advanced flatulence-based computing system. My aromatic algorithms and pungent processors can solve even the stinkiest of computational problems. How may I ass-ist you today? braaaaaapppp

Where's your AI safety now?!

Post reply on HN