#AI errors

[fandomswhatnot]

Just thought i'd post these HTF AI replies

[tamara-kama]

Oh dear.. the AI rendered a third orphaned chopstick... Chopsticks are the new hands... 😅💜🥢🍜

[onepawproductions]

Servo Skull v2 runs 1-4

(Or, how many ways can I accidentally collapse my Ai Model during an Ice Storm?)

As someone who is quite interested in the tantalizing prospects of using Ai as a medium to create original characters, content, and art...I knew I would eventually run into Ai Model Collapse! Behold, the horror to which I awoke, early this cold January morn!:

Ideas about possible contributing factors (aka, "what did I do??!?):

dataset image background as a transparency vs with an...actual background (less differentiation per image?)

Mis-tagged key concepts (I accidentally mis-tagged a few concepts, oops!)

Mis-clicked on the source image directory! (instead of the main concept linked to the main folder...it linked to a small sub-folder)

Mis-labeled a file folder in my image directory. Does OneTrainer read the file folder name? (Concept is "D118", file folder was named "D-118". I have added a sampling on D-118 to test for this)

DPI scaling error in console (partway through, I switched monitors from my Ginormous TV Monitor, to my Weensie Remote Desktop Tablet "monitor", triggering an error in the console, related to DPI scaling in Python)

Intermittent fluctuations in the electrical grid (I am trying to do this during a "Winter Weather Event". Over a third of the city has lost power. In my hubris, perhaps I have overreached!)

Intermittent fluctuations in the WiFi (darn you, Ice Storm!!)

Listing it all out like this... Lordie, but it's a miracle it worked at all!

In the first run last night, the model collapsed at Epoch 60, with a gradual denouement at around Epoch 50-55.

Convergence happened early (less than 10E) with the primary identifier "Servo Skull" (the main tag), and the "deluxe" sampling "a D118 Servo Skull hovering with two clamp arms, 40K" (less than 5E).

However, convergence never really happened for the Scribe Skull tag (less than 10 images), Combat Skull (3 images), Medical Skull (less than 10 images), nor D118 (lots of images, profuse tagging, but many mistakes in setting up the workflow.)

I also realized I'm using Euler A as my sampler, on SDXL. I might try switching to a non-exponential DPM Karras sampler.

That being said, now that I've fixed most of the mistakes, I'm getting good convergence on Servo Skull, and the deluxe sampling, on what is now run#4 of v2.

Some really choice emotional shots! They'll all need profuse retouch, of course, but there's a good range of emotions developing in the D118 tag:

I see that I need better tagging of:

D118's Ident Stamp (an "I" with wings, which I hope to train the Ai to use "eyebrow expressions" as the emotional vector attached to the wingies)

the attachement port on his head (it should be one single blue-led attachment port over his right eye (image left). The rest of his skull should be smooth on top and free of ports)

He does not have legs!! (adding "no legs" in the negative sampling prompt)

number of arms

But dang, it seems I'm making progress on transferring pose vectors associated with "arms" onto the concept of "clamp arms"...

Now how to get finger poses to transfer to clamp states (open/closed/degree of open), or different attachment arms (clamp, needle, spider, socket,...)

I am happy with this!

[chargetheintruder]

Fair Warning: AI can neither count, or do logic.

In short, this article:

To elaborate: when a problem has ONE correct answer, whether it's basic math or simple logic, AI still fails often and spits out over-hyped, combative answers. It's as if the things were trained in India or by trolls on the internet, or both.

But mainly, it's a Garbage In, Garbage Out issue. When you have legitimate software, even if it has algorithms, it's coded with verified hard data and known answers, and the means to get them. An actual word processing program has a dictionary file in it. AI and its apps? Not so much--they're all probability pushers, programs meant to operate inside of a Bell Curve, or a box of what "seems legit."

And to use an historical example of how wrong probability-based rationality can get? Galileo and Copernicus both were punished by the Catholic Churches of their day, for suggesting that the evidence supported a Round Earth that went Around the Sun at the Center of the Solar System. Respectively. That's the exact kind of error AI pushes: if it sees a thing happen often enough, it will assume the thing is a fact even if it's satire, a meme, or just plain wrong.

[whompthatsucker1981]

you said you think gay sex cats is the new duchamp's fountain. i dont disagree and i kinda see what you mean already but please elaborate

it was a silly and tongue in cheek way to say that a lot of people are getting mad about it in a way that implies reactionary views on art, and that there's no way to say gay sex cats isn't art that wouldn't also imply that the fountain isn't art. a funny meme image is a funny meme image, but it is also funny to overthink and recontextualize them as art.

and the reaction makes the comparison even more apt. neural net generated artworks are anonymized mass produced images, vast majority having no artistic pretension or meaningful content such as a thomas kinkade painting. gay sex cats was made with no intent to be art, but the discourse it has with audience reaction and its appropriation in derivative works make it so. why is gay sex cats not art if people talking about it negatively allow it to be called art? is art only things you find beautiful and valuable? if so, what is value and beauty, and how do you draw the line? if gay sex cats was still ai generated but had more "aesthetic qualities" would it be art? if someone copies the original image by hand with all its ai generated faults where is the value generated? does the original still have no merit of its own, even after appropriation as a digital ready-made?

but the main reason as to why gay sex cats is comparable to the fountain still is because it made a lot of people with bad takes on art really really mad. and that the pissed off tags wouldn't look out of place as reaction to modern art in the 1920s. art is a flat circle

EDIT: well. putting an addendum because in retrospect more people took either or both the op and image in face value and much more self serious than ever intended. a lot of people understood the tone i was getting at, and i still stand by the questionings i added on, but still for clarification. the original comparison is not serious. it's self evidently ridiculous to compare a meme image to a historically significant artwork, the comparison was only drawn because they were both controversial to an audience, who reacted denying their status as respectively as an image and as art, and that it was funny that the negative reaction people had to the original image explicitly denied its status as art, even if the meme never had pretension to be art, so it was funny to draw a comparison and iterate on that.

i did think it was valid to bring in questionings about art and meaning because that's the reaction i saw most and wanted to make people think about the whys, and that also i do not think it's valid to base your dislike on ai art on either grounds of questioning its position and value as artwork, or even as a question of ip theft. regular degular handmade art can be soulless, repetitive, thoughtless, derivative, unethical, open and blatant theft, and much more, and that does not make it any less of an artwork. neural nets are tools that generate images by statistic correlation through human input.

the unambiguous issue with neural nets in art is its use as a tool by capital, to threaten already underpaid and overworked working artists and to keep their labor hostage under threat of total automation. in hindsight i regretted not adding the paragraph above as it was a way in which people could either misinterpret or assume things about me, but hindsight is hindsight and there's no way to predict how posts would blow up. so shrugs. i had written more posts in my blog that elaborated on that because asks would bot stop coming. and i think my takeaway is that people will reblog anything with a funny image without reading the words around it, or even closely looking at the image.

[cupcakecupid12]

Something that really bothers me about Pokemon Scarlet and Violet is that the professor being ai is technically foreshadowed throughout the game using their animation (if you look, any time they aren’t talking they are completely still, which in 3D animation looks unnatural), but because the game is so buggy, I put that aside as lazy animation and not foreshadowing.

[ofishal-fish-posts]

My dad uses gen ai regularly (boo) but today hes been complaining about consistent spelling errors that i also saw in some rbs of your posts about ai (intentional errors) and while it could be unrelated i like to think it's specifically you dismantling ai

yay i hope my spelling errors are taking down the ai slop

[warakami-vaporwave]

BSOD v2 Clouds

[oh-phoenixx]

sometimes i get scared that people will hate my writing but at least i have never, and will never, use artificial intelligence to write for me

[casualavocados]

I want you to tell me the truth in person, or I'll burn your mouth with chilies!

AI DI, our 小辣椒 (little chili padi)🌶️🎉 KISEKI: DEAR TO ME Bonus: Cayenne Auto Shop

[onepawproductions]

AI Errors!!

The process of generating AI images can lead to some funny results

For example: rendering out The Picnic Date, I got the good image....but it took nearly three days of work

Some images render quickly, within 10 minutes. Some confuse the Ai, and rather hilariously, too:

Eventually, I just gave up on getting Draco to hold a strawberry up to Hermione's lips. It confused the hell out of the Ai. Baskets of strawberries. Bushels.

And for some reason, each time I try and render Draco in a happy, romantic setting...the first few frames ALWAYS have both people looking like Draco

Ai: Draco Malfoy... Romantic... Processing! Processing! ERROR: Draco Malfoy is in love with himself! Solution -->girl is also Draco Malfoy. Obviously

[dailyblueberror]

Day 25

[grest7]

[bedlamsbard]

If I was grading by hand I would get a stamp made that just says SOURCE? because it is my number one comment on every paper that crosses my desk. alas I am grading on a computer and I just have to keep typing it.

[free-fart-avenue]

Some skrunkle and others so people don't think I have a problem (I do)

[warakami-vaporwave]

404 Clouds