Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
not them adding extra skin to make it look like a hole in their face therefore m…
ytc_UgzKU_Jqx…
G
Historically, the West has utilized new technologies for military or imperialist…
ytc_UgxXSPi27…
G
China is using artificial intelligence to make advanced in technological warfare…
ytc_UgyS4Mgih…
G
All neural network AIs have inherent data biases. Any other company would have t…
ytr_UgxNN28vB…
G
No, AI, like any other piece of tech, does exactly what we tell it to do, and th…
ytc_UgzlNS7h6…
G
Hank, *please* read Adam Becker's "More Everything Forever". If you're going to …
ytc_UgwfbTS8P…
G
Only way is to regulate the amount of AI to human resource ratio in companies.…
ytc_Ugx2rRo9n…
G
Bhavik Sihmar Unless you work high up in AI development you have NO CLUE what yo…
ytr_UgzP4mdx9…
Comment
⚠️ What researchers are actually concerned about
The real takeaway from these studies is much more grounded:
AI can produce harmful or manipulative ideas if prompted incorrectly
So developers need:
better safety rules
stronger filtering
clearer boundaries
That’s why systems like me (CHATGPT) are trained to:
refuse harmful instructions
avoid manipulative or dangerous outputs
🚫 What the video gets wrong
The video jumps from:
“AI can generate bad ideas in a test”
to:
“AI will try to kill humans to survive”
That leap is not supported by the research.
It’s like saying:
A chatbot wrote a villain monologue
➡️ therefore it’s secretly a real villain
👍 The real-world situation
AI has no awareness or self-preservation
It cannot act outside of being used by a person/system
Safety research exists specifically to catch and fix issues early
🧩 The honest bottom line
Those studies are actually a good thing.
They mean:
“Scientists are stress-testing AI to make sure it behaves safely—even in weird situations.”
Not:
“AI is secretly plotting against people.
youtube
AI Harm Incident
2026-03-18T14:4…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | developer |
| Reasoning | consequentialist |
| Policy | regulate |
| Emotion | approval |
| Coded at | 2026-04-27T06:26:44.938723 |
Raw LLM Response
[
{"id":"ytc_UgyE2KRBw3iJYZUh7Fh4AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"regulate","emotion":"approval"},
{"id":"ytc_Ugwz_psbw3fbbCaVzi54AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},
{"id":"ytc_UgzYq74k2Lv4qFQcG3p4AaABAg","responsibility":"developer","reasoning":"virtue","policy":"unclear","emotion":"outrage"},
{"id":"ytc_UgzMrsfi4YbsP1_Yu8N4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytc_UgybeqOUonoYFaaRcf14AaABAg","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"ytc_UgyzqG0Y-oN5g-XKJq14AaABAg","responsibility":"company","reasoning":"deontological","policy":"unclear","emotion":"outrage"},
{"id":"ytc_Ugyy0-afwEJOJesnf_J4AaABAg","responsibility":"user","reasoning":"mixed","policy":"none","emotion":"approval"},
{"id":"ytc_UgwY3UK3eSF0ZluhNUV4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},
{"id":"ytc_UgwPKldGDkwIgfnosAR4AaABAg","responsibility":"government","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},
{"id":"ytc_Ugy6eO73zJSCrYNHKbJ4AaABAg","responsibility":"user","reasoning":"virtue","policy":"none","emotion":"approval"}
]