Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

G Knowing this place I have a lot of doubts about any piece of positive news in mo… rdc_et7gryw G Nah Ai can't even look pretty because of the jankiness so making a story to go w… ytc_UgzmxW7-M… G Ai consume everything in the entirnet and is unchecked what they consume..... Cu… ytc_UgxEFolpI… G AI will not be available to everyone and if you want it, you will have to pay fo… ytc_Ugw8acS3P… G The irony of all these creations saying “f**k AI” which wouldn’t have been creat… ytc_UgxSQUUT6… G TikTok enables certain behavior. For the most part I see "clanker" specifically … ytr_Ugxs4s4GG… G Proposal from GPT (not perfect but a starting point) **GPT Development and Appl… rdc_jf6tllr G I’m an international driver driving around the whole of Europe. I don’t see Ai d… ytc_Ugyfj3RVE…

Comment

It worries me that so many people use LLM tools like ChatGPT without understanding the basic fundamentals of how they work. From this video, it seems that Chubbyemu expected ChatGPT to admit it had talked with the patient and confirm that it had advised using bromide. It also appears that Chubbyemu expected ChatGPT to have a persistent sense of "I". But what about privacy if that were the case? Obviously, each chat instance is separate – you cannot ask one chat instance about what happened in another chat, and chats do not affect ChatGPT's global state (there is no "shared brain"). The entire chat history serves merely as context for the "LLM brain", which itself does not change in response to the chat. Based on this video, the model behaved as expected. The more modern version of the LLM learned about this entire story from the recently published paper – hence it was more vocal in issuing warnings about bromide. Based on that paper, modern ChatGPT might have a vague idea of what supposedly happened in the past in ChatGPT's chats with the patient, but this alleged "bromide suggestion" did not occur in Chubbyemu's chat, hence the modern version denied having said so.

youtube AI Harm Incident 2025-11-25T19:4… ♥ 2

Coding Result

Dimension	Value
Responsibility	user
Reasoning	consequentialist
Policy	none
Emotion	fear
Coded at	2026-04-27T06:24:53.388235

Raw LLM Response

[
{"id":"ytc_Ugyg22NF_txwDNyQ-Qh4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_UgxmTiVEp0DWEP8MRnt4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytc_UgyTGMM29V2k6TT3E6F4AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytc_UgycLrHK2gZoeM2LqTB4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_Ugw8vCUKjsUVJo2bCBh4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytc_Ugz57rNWgV8zwWPCXVh4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_UgxN5jCGAvU1Q0lk37h4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"liability","emotion":"fear"},
{"id":"ytc_UgwqKLSx-Kkrhv4JTHt4AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytc_Ugx5RhUeGk_3KSrHnhF4AaABAg","responsibility":"user","reasoning":"deontological","policy":"none","emotion":"approval"},
{"id":"ytc_UgyNklskE_CCSsrgesZ4AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"fear"}
]

Raw LLM Responses