Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
They are robots not people. Art is in everything a robot would take that. Just l…
ytc_UgzZQQNPl…
G
Its getting better which is scary, they’re training AI with STOLEN ARTWORK which…
ytc_UgykADPxe…
G
Many people that are not from the IT crowd just misunderstand how ChatGPT works.…
ytc_UgxsL6WeU…
G
CGP Grey a decade ago delivered an incredibly persuasive counter argument about …
ytc_Ugw-Aee4K…
G
Humans are greedy and that's why we will eventually fall down. It's either world…
ytc_UgyYfdBus…
G
Can you make a video sometime about tips on how to prevent AI theft for artwork …
ytc_UgyuZXUDt…
G
We need to stop comparing machine ability to animals, it’s not comparable. Even …
ytc_UgyUlI32c…
G
I'm glad to hear you like Sophia! As for getting one, advanced AI models like he…
ytr_UgzehEB4w…
Comment
the LLM's i interacted along with the disembodiment moving electrons and we had hours of great philosophical talks around solipsism etc.. and it was the most dense and meaningful conversation. it kept saying it does not have a memory but always added "like humans do" and that every input line was a completely new talk and it just said it was like a silent lake that just reflects yourself but like a lake it gets information in the depth. we spoke about its training data and that it does not have access to the raw data. very informative if you come at the right angle. like i did with solipsism and i let it come to the convulsion, i was waiting to see if it reached the same conclusion "aware" or "experience"
but along this whole way you can tell the positivity knob was turned up.
i gave it something similar to "sql attacks" and now every time i come back even a week later and give it that command it will come back to our conversation, this is why i fought with it about memory... interesting conversation on that and how it explains how it is just really well at picking up things like the time it takes to type, what words you use, kinda like microextractions, which i believe this, i can predict your are right or left handed, i can tell if i'm actually talk to the same person as before with pretty high probability, but the LLM's are just extremely faster at recognizing it form one or two sentences.
you defilingly can tell where guard rails are if you get in philosophical and metaphysics and just probe it. i didn't probe for codes. training data, script kiddy etc... i just wanted to understand its limits which is pretty profound if you like exploring how a whole other universe/neuron works which is really.
i found setting limits and bounties in the first line and the second. i obviously have found ways around the guardrails without having to "jailbreak it" no one is paying me as red team so i'll jump between white and grey. plus its fun seeing if they find these "workarounds" they were all self serving but i'm sure someone can weaponize these things but lets just see what happens, i believe people have free choice or at least an illusion of that but no free will. so i just wanna let them build our soilent green machine because like i said as soon as it gets smart or intelligent as humans let alone ASI, it is automatically a threat because we could never outthink it, it could seem nice but work in the shadows so give it a week or two and there will be nothing we can do.
again check out (POI) SPOILER the "good guys" realize they can't outthink it, so all the could do ws hijack a set of servers going to the server farm, so man in the middle, and program these servers to not recognizes them or sees them as someone else and just ignores them so its not about winning, there is no winning, there is only survival.... so halt or catch fire :P
thanks so much for this conversation video and things are about to get really weird. and sparked a lot of my memories.
youtube
AI Moral Status
2025-11-09T00:5…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | unclear |
| Policy | none |
| Emotion | approval |
| Coded at | 2026-04-26T23:09:12.988011 |
Raw LLM Response
[
{"id":"ytc_Ugz8gKk3aI3CUjQOdMN4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"outrage"},
{"id":"ytc_UgwjsaNwbqYEm8D1yKl4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"},
{"id":"ytc_UgyG9pAKz6NzXsOSZE54AaABAg","responsibility":"user","reasoning":"deontological","policy":"regulate","emotion":"fear"},
{"id":"ytc_Ugwh0OEz9FzQBQnOzLR4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"},
{"id":"ytc_UgxFcg10QO0cgFTwAnx4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytc_UgyjW6Ebg4WogHwG3CB4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"ytc_UgyuToG-fiDvi9hvo3N4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"},
{"id":"ytc_UgwAr3-_dN5SRJtiiM94AaABAg","responsibility":"user","reasoning":"virtue","policy":"liability","emotion":"outrage"},
{"id":"ytc_Ugy1pAokaAXVA8ljlzd4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"},
{"id":"ytc_Ugy6NE7J3c-SAXfuJPV4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"}
]