Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
AI art is just an efficient way of making animation series. There's no talent in…
ytc_UgyzMS9-p…
G
Great video.
Same with coding, the boilerplate will be provided by the IA, but t…
ytc_UgzJ9VnyU…
G
He is talking nonsense about risks with AI ! And that gives not a good reputatio…
ytc_Ugw7A_VuP…
G
There are two counterposed goals in the design of chatgpt - one is to sell their…
ytc_UgxAiZS8P…
G
AI created....RISK, Human created ....RISK, Nature created.....Risk. Which Risk …
ytc_UgxDOay1Y…
G
This was probably the worst AI video i have ever watched. This person is just ap…
ytr_UgwQur6ht…
G
To all of those Trump supporters watching this, he just got a shout out in this …
ytc_UgxG-V4eF…
G
AI became the definition is bastardized by morons. It is well different from se…
ytc_UgwQ0BLBf…
Comment
At 45:29 Eliezer tells a story with a rhetorical trick that clouds the very argument he is making. He speaks of AI killing all of humanity, and humanity fixing AI so it won't happen again, and it happens again, and we fix it again, and the pattern repeats. Then he points out that AI only has to kill all of humanity once, and humans can't learn from our own global extinction.
But species extinction in the natural world is a better analogy and more nuanced than what was discussed. Natural selection has shaped every species and extinguished many over millions of years. On the other hand, in a few hundred years, humankind has created conditions that have caused massive extinctions. But there is a recognizable pattern to these human-caused extinctions and a series of steps prior.
We need to learn how to identify and respond to the many warning signs that will come well before an AI-driven extinction. And we need to structure the releases and free movements of each AI so that we have many opportunities to learn and fix.
youtube
AI Governance
2025-12-01T19:5…
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | consequentialist |
| Policy | none |
| Emotion | mixed |
| Coded at | 2026-04-26T23:09:12.988011 |
Raw LLM Response
[
{"id":"ytc_UgwfuJldpu13N5yIjgJ4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytc_UgwQtPiShjExt_Mm1Vp4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytc_Ugzd-yLBfi9WMHa4g0p4AaABAg","responsibility":"none","reasoning":"mixed","policy":"none","emotion":"resignation"},
{"id":"ytc_UgwnLFQDQY47429ZVih4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"mixed"},
{"id":"ytc_UgyENebu1tHpusFjJtd4AaABAg","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"ytc_Ugz0siumGK2Szqinj4x4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"indifference"},
{"id":"ytc_UgwDEYCalO3RoZEuB_J4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"ytc_UgzdMVHYcOIlKj399Gl4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytc_UgzsQj-ugSeNf558p_d4AaABAg","responsibility":"ai_itself","reasoning":"consequentialist","policy":"none","emotion":"resignation"},
{"id":"ytc_Ugw-wjLHTVGqJ0RSS-t4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"mixed"}
]