Raw LLM Responses
Inspect the exact model output for any coded comment.
Look up by comment ID
Random samples — click to inspect
G
I feel a similar thing is happening with online articles. I keep coming across …
rdc_lkc2o34
G
watching deepfake porn of your friend that was made without their consent is wei…
ytr_UgwPm8HXy…
G
AI can’t ever be 100% human, just like humans will never be 100% AI. It just can…
ytc_UgwQZlOHF…
G
@jpnewman1688The humans building AI think they are gods. So their reasoning is …
ytr_UgxIrMQwN…
G
A) its AI systems “listening,” not actual humans.
B) they’re trying to sell you…
ytc_UgyBTL_Lg…
G
I think this dude just has nudes and stuff from her thats why she is crying caus…
ytr_UgwcMFVif…
G
I am so glad I started therapy and transitioning from republican cishet world be…
ytc_UgxSxcytb…
G
God Father Of AI 😂😂😂 China Is Decades Ahead Of All Western European Countries Yo…
ytc_UgzOiKAgm…
Comment
Your analogy of AI being a poopy baby without training is hilarious and true, and that it can't make anything without trainings from other artworks or photos, but this is no different than what all artists have done since the beginning of time.
If I go back in time to the 1800's and ask the world's best artist to draw me a car, they're going to have NO referential concept of what that is supposed to look like. Even if I describe in GREAT detail what it's supposed to look like, they're only going to be able to come up with something based on current conceptualized ideas of things they already know.
If I set down a pair of Apple Airpod Max's and you take them, that's theft. If I set down a pair of those headphones and you pick them up, analyze their construction, how they work, the proportional ratio of how they would fit onto someone's head, and reverse engineer a completely different pair of headphones, is that theft? I'd say no, in every way. This is how AI works, except instead of taking this data from one type of headphones, it takes data from ALL types of headphones to find the statistically highest probably match for the best outcome to create the new headphones.
Every single object on this earth has some sort of mathematical predictability. It's why all cats look like cats despite their breed, or all humans look like humans despite their race. These training models take the data of how those things are constructed and creates mathematical concepts for it to generate from. The odds of the neural network taking something uniquely distinct from YOUR pieces, that does not exist anywhere else from any other artwork, from the millions or even billions of photos scanned, is practically impossible, let alone a large enough piece to even remotely be considered "stealing."
Is it stealing to draw your character's eyes the same way another artist draws theirs because you're influenced by that? Did you ask that artist for permission to visually scan over their artwork and draw eyes in the same way? Do you think that no other artist completely unrelated to that artist didn't also draw their eyes in a similar way? This is the kind of over sensationalized hype and fear around AI that's absolutely ridiculous.
Furthermore, the "in the style of" argument seems fair, but is also limited in it's understanding. Van Gogh or Picasso were NOT the ONLY artists from that time period that made art that way, they were simply the most famous to do so. As you said, you can't copy an art style, but these artists being the most famous, or have produced the most referential artwork are the ones that provide the most training data. You can get almost the same results just by referencing things like "impressionism" or "cubism" or whatever other style you're looking for.
Regardless of all these subtle talking points, ALL creative forms of expression are learned, developed, and advanced from imitative expression. It's simply how the human mind works, to observe, and imitate, until the point you have enough observations from different influences to create something new.
youtube
2024-02-04T15:3…
♥ 3
Coding Result
| Dimension | Value |
|---|---|
| Responsibility | none |
| Reasoning | virtue |
| Policy | none |
| Emotion | indifference |
| Coded at | 2026-04-27T06:24:59.937377 |
Raw LLM Response
[
{"id":"ytc_UgyzU9KwNK5cTv-dQVB4AaABAg","responsibility":"none","reasoning":"consequentialist","policy":"none","emotion":"approval"},
{"id":"ytc_UgxEVQ2dn_jJt-tXSD14AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"indifference"},
{"id":"ytc_UgwdhWjKw7IPiNyY1cJ4AaABAg","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"ytc_UgwDzOr2Kh_9XnkLqOR4AaABAg","responsibility":"user","reasoning":"consequentialist","policy":"none","emotion":"fear"},
{"id":"ytc_Ugw7LhylY6MfS0EIeIt4AaABAg","responsibility":"distributed","reasoning":"contractualist","policy":"regulate","emotion":"approval"},
{"id":"ytc_Ugx5V1Bl2TZrQbYUHeN4AaABAg","responsibility":"none","reasoning":"virtue","policy":"none","emotion":"indifference"},
{"id":"ytc_UgyS63THOHVfM0hJq0B4AaABAg","responsibility":"company","reasoning":"deontological","policy":"liability","emotion":"outrage"},
{"id":"ytc_Ugw1ZTCwVBIH-nuPMVF4AaABAg","responsibility":"company","reasoning":"deontological","policy":"regulate","emotion":"outrage"},
{"id":"ytc_UgwRXft2tiGseUlsKQR4AaABAg","responsibility":"none","reasoning":"unclear","policy":"none","emotion":"approval"},
{"id":"ytc_UgwOU1ObDJeUCsVrZIF4AaABAg","responsibility":"ai_itself","reasoning":"deontological","policy":"none","emotion":"indifference"}
]