Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
The “Godfather of AI” comments on Elon Musk made the viewing of this episode dif…
ytc_UgzeYihSI…
G
I really dislike this situation I see ai art as being a possibly good or even fu…
ytc_UgxXouprg…
G
I knew I could do this I tried to jailbreak chatgpt but I couldn’t break through…
ytc_Ugy8ZRUB9…
G
@annaa3772I realize that is one of Brandon’s appeals to why not to use it, but i…
ytr_Ugymj3n78…
G
HOPE EXIST : In a reality so diverse and rich of options, this resolute surrende…
ytc_UgzbTGmjt…
G
No, it absolutely wouldn't render the argument null. Searle knows what he is sug…
ytr_Ugh9zENUy…
G
so that no one has anypower only Nvidia and the others. It`s real fracked howe …
ytc_Ugw8WLuc4…
G
A.I. is stealing from others who put their heart, effort and time into a peace o…
ytr_UgxZEkQMc…
Comment
It's even worse than that. AI Safety researchers predicted ahead of time that AI would scheme, self-preserve, and seek power, even before they knew what the architecture would be or how it would be trained. They knew this because doing those things isn't a property of humans; it's a property of goals.
Many current AI systems are agents, meaning they behave as if they have goals, but we can't robustly control what those goals are.
If something has a goal, almost no matter what the goal is, there are specific instrumental subgoals that are always useful. Like "keep existing," "gain resources," and "gain power." So even if we somehow made its training data squeaky clean and good and moral, when it is clever enough, it will still independently discover useful strategies that aren't what we want it to do.
Check out AI Safety Info if you want a more in-depth explanation, or take a look at PauseAI if you want to help steer the future away from a cliff!
youtube
AI Moral Status
2025-06-06T07:4…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | ai_itself |
| Reasoning | consequentialist |
| Policy | unclear |
| Emotion | fear |
| Coded at | 2026-04-27T06:24:59.937377 |
Raw LLM Response
[
{"id":"ytr_UgyCLLm0FDNKmLQxzuN4AaABAg.AIxmO9qRfR5AIzaKexHFS6","responsibility":"ai_itself","reasoning":"unclear","policy":"unclear","emotion":"fear"},
{"id":"ytr_UgxPBLoTKKoViW30UkR4AaABAg.AIxiTVeOZGGAJ-B2vZzwXI","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytr_UgyNLTnluXfxGwQ-NGR4AaABAg.AIxbb-fp30qAKT6UdVRo1E","responsibility":"unclear","reasoning":"consequentialist","policy":"unclear","emotion":"resignation"},
{"id":"ytr_UgzuU0OpWQvT_U5N4rJ4AaABAg.AIxYMBFoRceAJ0nswhTioR","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},
{"id":"ytr_UgzuU0OpWQvT_U5N4rJ4AaABAg.AIxYMBFoRceAJ2LCxLs8r-","responsibility":"developer","reasoning":"unclear","policy":"unclear","emotion":"outrage"},
{"id":"ytr_UgzuU0OpWQvT_U5N4rJ4AaABAg.AIxYMBFoRceAJZTYqEnGaf","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"approval"},
{"id":"ytr_UgyYnZttPQg9-QvAKNx4AaABAg.AIxY17JXIqiAIxf4hxEFTA","responsibility":"ai_itself","reasoning":"unclear","policy":"unclear","emotion":"resignation"},
{"id":"ytr_UgxCOML_yw6tpD0Iu5V4AaABAg.AIxXBHzBGkfAIxbNKQStBp","responsibility":"ai_itself","reasoning":"consequentialist","policy":"unclear","emotion":"fear"},
{"id":"ytr_UgxCOML_yw6tpD0Iu5V4AaABAg.AIxXBHzBGkfAIxe413NehS","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"indifference"},
{"id":"ytr_UgxCOML_yw6tpD0Iu5V4AaABAg.AIxXBHzBGkfAIyD-HSoL2V","responsibility":"unclear","reasoning":"unclear","policy":"unclear","emotion":"approval"}
]