Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
@laurentiuvladutmaneaif something is transformative then it is not stealing, so …
ytr_UgxFtCdev…
G
Using polite language appears to be the secret sauce in enticing the model to br…
ytc_UgzYjCwRz…
G
The AI agents we use current wrote upgrades themselves and some says don’t vibe …
ytc_UgzUPqeUH…
G
AI needs regulated. Trump and billionaires are only about $$$. The Big Ugly Bi…
ytc_Ugzdwqa9R…
G
heating premade food up in the microwave dont make you a chef. same with AI " ar…
ytc_UgzskEjid…
G
As a creative, I don't think that AI will take away all of the jobs of creatives…
ytc_UgxVZA-kv…
G
I definitely think that ai can be interesting, just not by itself. well, it is a…
ytc_UgzQfxq-K…
G
Three things: First, crossing the road, on a dark road, without any reflective g…
ytc_Ugx61-v8w…
Comment
So... 35GW, is nowhere near 1% of global energy consumption. We're currently producing and consuming about 20TW. 1% is 0.2TW, or 200GW. So it's 0.175% if it's 7, and 0.125% if it's 5. And considering there's already a 2.5GW datacenter (aws northern virginia) and a number of 1GW ones, it's less of a jump than you might think.
That being said... The infrastructure to supply 5GW per site isn't likely to exist. So that's going to have to be built. That's an investment for sure but also quite risky since the AI boom as it stands could pop at any moment and then that infrastructure isn't just a wasted investment, but it'll literally be detrimental to the grid in the area. So IMO, they should either have a completely seperate energy production off grid that supplies their needs (not likely), or they should be paying up front for their expected energy for a number of years as determined basically by how long it would take to fix the grid if they go away.
Large consumers already pay some of that cost for a couple of years ahead. But that should be way WAY extended for cases like this.
reddit
Cross-Cultural
1727486936.0
♥ 2
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | unclear |
| Reasoning | utilitarian |
| Policy | unclear |
| Emotion | indifference |
| Coded at | 2026-04-25T08:33:43.502452 |
Raw LLM Response
[
{"id":"rdc_lp9p32n","responsibility":"developer","reasoning":"virtue","policy":"none","emotion":"outrage"},
{"id":"rdc_lp9vw8m","responsibility":"unclear","reasoning":"consequentialist","policy":"unclear","emotion":"indifference"},
{"id":"rdc_lp9zjcw","responsibility":"unclear","reasoning":"consequentialist","policy":"unclear","emotion":"indifference"},
{"id":"rdc_lpau5n6","responsibility":"developer","reasoning":"virtue","policy":"none","emotion":"outrage"},
{"id":"rdc_lpbfg2v","responsibility":"ai_itself","reasoning":"unclear","policy":"unclear","emotion":"fear"}
]