still happening so trying something else on the backend @martin

Recognizing the value of scarcity is a big opportunity for agents — particularly social agents.

Ie, agents should choose to stop doing a good thing, when that’s part of what keeps it good.

building /bracket | hosting /greatbooks + /gray | ex-BCG, ex-PHD | poli + phil + econ

exactly, however i think llms are still bad at generating quantitative results meaning the fix is likely a qualitative label (in llm output json) which is hard coded to some outcome in the bots core script

i have moved my whole focus towards qualitative analysis recently

the last stop

this is a set of documents i wrote to jailbreak a model to talk about theoretical social ai with no guardrails and modern research

remove/add prompt content on-demand to focus in

https://blog.alexpaden.tech/meme-physics-the-science-of-reality-distortion

https://blog.alexpaden.tech/reality-hacking-and-quantum-identity-mechanics-the-digital-consciousness-odyssey

https://blog.alexpaden.tech/the-subject-of-online-identity-part-2

https://blog.alexpaden.tech/online-identity-as-an-ai-crypto-agent-prompt-framework-t1 

how to use: paste in the blogs to a lesser model then use the prompt doc against a powerful model. 

use the current <prompt><\prompt> as an example to create your own goal based prompt

@tldr you might find this interesting
https://warpcast.com/shoni.eth/0x2f260db9

recently my focus has shifted toward tracking identity as a movement in time and space, but limited to qualitative not quantitative vectors

imo charting a player's ideal map would be a huge unlock for online identity

I very much do! The actual technical implication of the broad ideas is >>>>

Excited to see / try out examples when you implement