Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
You’re right on the money with this video. I’m the author of prkl-ann, a C++ fra…
ytc_UgwJP3siM…
G
Tell us about the actual safety problems. Deleting the technology because it tak…
ytc_UgxKs7fV7…
G
At my big tech company people put 10-15k+ LoC in single files. That way AI agent…
rdc_n3kk6lp
G
Cameras suck at dark things. It’s been known for a while facial recognition has …
rdc_jv5to9o
G
I would be quite amazed that channel called "Documenting AGI" would do video lik…
ytc_UgyQMKL75…
G
"Predictive Policing" I mean, can hardly be mad at the AI, it's like giving a ch…
ytc_Ugw24fLnr…
G
Ask yourself why your not A I. You won't be able to convince yourself let alone…
ytc_Ugy1zofO2…
G
Yea yea, don't we know from where does AI get it's data and what to do?
People b…
ytc_Ugw3kKe7b…
Comment
I wasnt super worried about exisitential risks to humanity of the newest AGI contenders until I've seen the massive push towards Agentic AI happening across the entire tech world over this year. I really don't want LLMs or any future deep reasoning model to run or be able to interact with essential infrastructure, but i guess with the public release of MCP that train's left the station for good. Everybody and their mother wants to build agents now, may god help us when power systems become vulnerable to prompt injection.
I'm still grasping onto hope that AGI (or superintelligence I guess) is a lot harder to crack than what recent progress would suggest since from what I can tell, the raw performance improvements of LLMs this year have slowed down and energy needs grow astronomically, which would not be helped at all if the AI crash does happen eventually as by my understanding modern LLMs have not had a return of investment yet.
But what do I know, unlike Soares I haven't studied AI for many years, and from what little I know, he truly is an expert in the field. So I will take his concern in the matter seriously.
youtube
AI Moral Status
2025-11-06T23:5…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | developer |
| Reasoning | consequentialist |
| Policy | liability |
| Emotion | fear |
| Coded at | 2026-04-26T23:09:12.988011 |
Raw LLM Response
[
{"id":"ytc_UgzvBR3xx6FYttlkxYt4AaABAg","responsibility":"government","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"ytc_UgwQSTTf7v2pVbA_xQV4AaABAg","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"mixed"},
{"id":"ytc_Ugz2BPndcHlNRgzcuZl4AaABAg","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"indifference"},
{"id":"ytc_Ugxfn1OzBs-TIQjcWXJ4AaABAg","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"mixed"},
{"id":"ytc_UgzHTHA9O3SBJgyUy4J4AaABAg","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"indifference"},
{"id":"ytc_Ugx0GDbnzyQM4Vdlb7x4AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"ytc_Ugy7OHLiO9WMY1At9Dp4AaABAg","responsibility":"developer","reasoning":"virtue","policy":"none","emotion":"mixed"},
{"id":"ytc_UgzZhAxON4WPIXn7HDR4AaABAg","responsibility":"distributed","reasoning":"consequentialist","policy":"regulate","emotion":"fear"},
{"id":"ytc_Ugw_DQynDVHjJvl2m9t4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"unclear"},
{"id":"ytc_UgywzpC9VikcdWfMXuV4AaABAg","responsibility":"developer","reasoning":"consequentialist","policy":"liability","emotion":"fear"}
]