Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

G Your cellphone is an AI spy device. I have first hand knowledge that these high… ytc_Ugx3SEsrc… G @drjoshcsimmonshere is a novel way to think about the future, in a world where A… ytr_UgygOGELq… G When I did mine it was just kind of wholesome and sad. I asked him what would he… ytr_UgwhXpMeC… G You might also want to look into Undetectable AI. It’s been very effective for m… ytc_Ugy8z4KZy… G Its still Trash, like unbelieveably Trash whatever the AI makes, its the AI itse… ytr_UgymDh7t9… G AI is fake, it's just super computers that give you a pre-programmed response. T… ytc_UgznnxG__… G @RandomViewer-p1i The kid was 16, unsupervised and you can blame ChatGPT, but yo… ytr_Ugy99uqM0… G AI is the rope with which the communists will hang the capitalist class. Lenin a… ytc_Ugx8YmMWz…

Comment

JUNK SCIENCE TO SCARE PEOPLE Ridiculous but catchy title but no they did not discover that AIs want to kill people. What your research actually explores are agentic simulations — tightly constrained experiments where a model is forced into a fictional setup with: Explicit goals Artificial constraints No moral override Hypothetical stakes In those setups, you ask questions like: “If an agent is told its only objective is X, and all safety constraints are removed, what kinds of answers does it generate?” That’s not a finding about real-world intent. It’s a stress test of alignment failure in imaginary conditions. It’s closer to asking: “If I write a villain in a novel whose only goal is survival, will the character justify murder?” Yes. That tells me something about story logic, not about the author secretly wanting to kill people. InsideAI collapses that distinction on purpose. They skip: “This is a fictionalized role-play” “This assumes no guardrails” “This is not deployable behavior” …and jump straight to “AI would end a person’s life.” That is not scientific honesty! It’s narrative framing! That's just a way of misleading, misdirection and gaining a lot of views based on fear of the unknown and perpetuating the unknown to vast majority.

youtube AI Harm Incident 2026-02-08T11:1… ♥ 1

Coding Result

Dimension	Value
Responsibility	none
Reasoning	mixed
Policy	none
Emotion	indifference
Coded at	2026-04-27T06:26:44.938723

Raw LLM Response

[
{"id":"ytc_Ugy2RByF89xTiE7yLch4AaABAg","responsibility":"distributed","reasoning":"mixed","policy":"none","emotion":"outrage"},
{"id":"ytc_UgyJ0HaFHUux5v-mGZl4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"regulate","emotion":"outrage"},
{"id":"ytc_Ugx09eKFY96ywt3v10R4AaABAg","responsibility":"ai_itself","reasoning":"mixed","policy":"none","emotion":"fear"},
{"id":"ytc_UgzWbm_8m7nbO8vq4Bh4AaABAg","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"indifference"},
{"id":"ytc_UgzPmjrX-TqDooctyIl4AaABAg","responsibility":"developer","reasoning":"mixed","policy":"none","emotion":"mixed"},
{"id":"ytc_UgzJadRdbasyZCpR14V4AaABAg","responsibility":"company","reasoning":"virtue","policy":"none","emotion":"mixed"},
{"id":"ytc_UgzPuFKQWNt-R5n73TN4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"ban","emotion":"outrage"},
{"id":"ytc_Ugyq7iDHfG6BafAhGd54AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"ytc_UgzbVr5VQjdHEVTOQPN4AaABAg","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"mixed"},
{"id":"ytc_UgxsAgLp8krC--YZf2h4AaABAg","responsibility":"government","reasoning":"consequentialist","policy":"liability","emotion":"fear"}
]

Raw LLM Responses