Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
both are made with ai but to trick us. one was generated with ai the other one e…
ytc_UgzpTcvmj…
G
I’ve been drawing since I was a toddler and I’m 16 now and still ass lemme just …
ytc_UgxYqpiV_…
G
They will out think us especially if we are in a stressed out position. Just the…
ytc_UgxtbANoZ…
G
IsRawHell is the father of AI!
They have the supreme power in cyber space!
So ju…
ytc_Ugzz1lXKq…
G
That is a chillingly accurate assessment of the "Architecture of Control." You’v…
ytc_Ugx929kf5…
G
“AI misalignment” is 100% different from “AI misinformation”.
What’s going to ki…
ytc_UgyguxlSm…
G
So the AI is smart enough to literally outmaneuver the entire human race to the …
ytc_UgxSU7rdy…
G
a robot may save your life someday ...either on the side of a cliff or in a nurs…
ytr_UgwHNvxlc…
Comment
Great interview, but at times quite disingenuous. Let’s not pretend OpenAI invented the scaling paradigm; they were simply the first to apply it decisively to language models and follow it through to its logical conclusion. And it’s hard to argue that this approach wasn’t spectacularly successful, at least in the early years. After all, the only real difference between GPT-2, which is barely coherent, and GPT-3, which triggered the global race to AGI, is scale, since both ultimately rely on the same transformer architecture.
And scaling works more broadly than just for language models. Richard Sutton called it the bitter lesson: to paraphrase, throwing more compute at a problem tends to outperform hand-crafted, domain-specific approaches over time. The idea that one could achieve comparable performance or range of capabilities by training on small datasets using few chips is simply not credible. It’s not even in the same league as what modern large language models can do.
youtube
Cross-Cultural
2025-07-11T14:4…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | company |
| Reasoning | mixed |
| Policy | none |
| Emotion | indifference |
| Coded at | 2026-04-27T06:26:44.938723 |
Raw LLM Response
[
{"id":"ytc_UgzdRXsJFV2xDkq4Vg14AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"outrage"},
{"id":"ytc_UgxLM9l8Wb5i-_gCB7F4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"outrage"},
{"id":"ytc_UgyxzMuxBGjG7FtszNp4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytc_UgzDcZ_JcGfuHaO_9cZ4AaABAg","responsibility":"company","reasoning":"mixed","policy":"none","emotion":"indifference"},
{"id":"ytc_Ugz11tMYtVPjHsToUZN4AaABAg","responsibility":"government","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_UgyAVVR5KUJ9SNffTZF4AaABAg","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"outrage"},
{"id":"ytc_UgzEurs5anu9OM_iNwd4AaABAg","responsibility":"none","reasoning":"deontological","policy":"none","emotion":"outrage"},
{"id":"ytc_UgzzHU0u0uyfMXlFKpB4AaABAg","responsibility":"company","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytc_Ugw3F9HXEPjNpqQTSAd4AaABAg","responsibility":"none","reasoning":"deontological","policy":"regulate","emotion":"fear"},
{"id":"ytc_UgwXkovWOzRFycK7TwF4AaABAg","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"outrage"}
]