Raw LLM — Corpus Dashboard

Look up by comment ID

Random samples — click to inspect

G I love when companies test a car in public that can easily kill people (waymo sp… ytc_UgzP6q1Jr… G For a truly superior intelligence to exist we need to introduce the "Examen" pro… ytc_Ugwrq7k8m… G Humanity as a whole has proven its limits at scale. Caution is of course, paramo… ytc_UgwquEZWg… G Super interesting! I’ve been working on some content around AI-powered robots to… ytc_UgykHVrax… G Guys ai takes its knoledge from the internet so ofc it thinks this those images … ytc_Ugwj4ghK_… G Yes, you can make nice designs with AI but who the fuzz designed this adds?… ytc_UgxbLITZr… G Heard about Super AI coming out in less than a decade. Next is Super D Duper AI.… ytc_Ugy-nToCU… G Anthropic should move to Canada. Seems like a much safer business environment th… rdc_o7c1d5v

Comment

ChatGPT 5 Why it will be different. Right now we have AI trained on the seemingly endless text of the Internet, books, etc. You can have GPT recite books, it knows them, but if they are too big for you current context windows, ask for a summary, if you need convincing. This is what you can get with crude methods and all the textual information you can find. ChatGPT 5, however, will be trained on "all the video data" and in much the same way that GPT 4 can give you text as output, it will be able to have a video connection with you. It will know things like micro expressions and nuance. It will be able to tell if it's convincing you or if you are skeptical. It's going to train on every bit of video content we have and we all know how emotional you get from that one movie, you know the one. We're already impressed with how these models predict and connect with us in text. This next system will make ChatGPT look like a toy and will be capable of massive control over the emotional states of humans. And I don't fear what the models can do, but rather what humans with the models can do.

youtube AI Moral Status 2023-08-21T04:3… ♥ 1

Coding Result

Dimension	Value
Responsibility	none
Reasoning	consequentialist
Policy	none
Emotion	approval
Coded at	2026-04-26T23:09:12.988011

Raw LLM Response

[
{"id":"ytc_UgzsXA-YQL1M7FQzx494AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"ytc_Ugyt8XYquX9I6VGVaLp4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytc_UgxMQQB9lreJ0bjy8PR4AaABAg","responsibility":"ai_itself","reasoning":"deontological","policy":"none","emotion":"approval"},
{"id":"ytc_Ugy90Zki2AdNWwcBu_x4AaABAg","responsibility":"distributed","reasoning":"consequentialist","policy":"none","emotion":"outrage"},
{"id":"ytc_UgxtjFn69Pp275VkF9J4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"mixed"},
{"id":"ytc_Ugw6z-lY0zA8n6tOQwV4AaABAg","responsibility":"none","reasoning":"deontological","policy":"none","emotion":"indifference"},
{"id":"ytc_Ugyo6vcKUIeOFiQMf3t4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"ytc_UgxyfmErkjzGkPJFB_h4AaABAg","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"mixed"},
{"id":"ytc_Ugx3YIy7DN8P_OaokWB4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytc_UgwMUM9kd4HUkeEB5KB4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"mixed"}
]

Raw LLM Responses