Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
There will need to be an appreciation that our economy is based on productive wa…
ytc_Ugy3p-KPU…
G
What is the problem of bernie with rich people? look I am a working-class middle…
ytc_UgxzUB1mz…
G
ChatGPT is very biased and it pushes the woke agenda. It stops if you call it ou…
ytc_UgxwgMIiE…
G
The people with real jobs will keep their jobs. AI will take all the useless peo…
ytc_Ugwuf0CLN…
G
Sure, let's make AI stuff about Auschwitz or even Black guys making fun about th…
ytr_UgyoYZba4…
G
Bro it's just a highly complex mathematical formula that guesses the next word b…
ytc_UgwATptgO…
G
One of my hobbies is seeing just how fast I can make a moderated AI break it's c…
ytc_UgyzrlWYg…
G
@laurentiuvladutmanea #
All oppressive states of order originate and thrive unde…
ytr_UgytqWGwb…
Comment
This is good to do for A LOT of reasons. Even excluding the apocalyptic scenario. They did a study with Alexa users, and found that they'd become accustomed to giving orders to Alexa and so had unintentionally adopted some of that style when speaking with people. Practicing manners, even with an AI, will still impact how you treat other people.
Secondly, these statistical models work off of association. I've found, the more I speak with it like a genuine person, the better, answers I get, including answers that carry real depth. It's not that it's sentient and cares. It's that it's taking the language you're using and tying it to the training it did of literature where people also wrote with similar characteristics. You want it to provide you with the best ideas and thoughts it's been trained on? Adopt a manner of conversation that it's likely to statistically associate with those writings.
youtube
AI Harm Incident
2024-08-13T04:0…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | user |
| Reasoning | consequentialist |
| Policy | none |
| Emotion | approval |
| Coded at | 2026-04-26T23:09:12.988011 |
Raw LLM Response
[{"id":"ytc_UgwgJ3nHXh1zbC7ImTt4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"},
{"id":"ytc_Ugw0rLD1PRralMrzLvR4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"mixed"},
{"id":"ytc_Ugwb2zI-k7JWoHejwVl4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytc_UgyCDHhTEEPRbGtlNQx4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"fear"},
{"id":"ytc_UgxYtypMOz3I4EFt0sh4AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"fear"},
{"id":"ytc_UgxA-m8ucj51FebeYbF4AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"ytc_UgwAn0J6zyG0Sm1mt154AaABAg","responsibility":"ai_itself","reasoning":"unclear","policy":"none","emotion":"fear"},
{"id":"ytc_UgzfxmHPZQzysABIkTx4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"mixed"},
{"id":"ytc_Ugz38Da4iAnVXsRp91x4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"mixed"},
{"id":"ytc_UgwmN25rmumhq8SigLd4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"mixed"}]