Thread #108213122
File: 138713513_p0.png (1.7 MB)
1.7 MB PNG
I simply do not give a shit about AI.
Discuss not-AI technology.
55 RepliesView Thread
>>
>>
>>
File: web.png (163.8 KB)
163.8 KB PNG
>>108213122
This is the best thread of 2026 so far
>>
>>
>>
File: 1girl, standing, large breasts s-2783106245.png (1.3 MB)
1.3 MB PNG
NovelAI general? It's about time.
>>
File: 1755372990304038.png (1.5 MB)
1.5 MB PNG
>>108213122
>NEW TECHNOLOGY BAD
>FIRE BAD IT HURT MY FINGERS
>SUN IS ENOUGH TECHNOLOGY
>DISCUSS THE NOT-FIRE TECHNOLOGY
>>
File: 8594455273_23f87d3daf_b-2354212267.jpg (73.2 KB)
73.2 KB JPG
>>108213153
>That's like asking people not to think about a purple elephant.
What? No one even talks about PHP anymore.
>>
>>
>>
>>
>>
>>
>>
>>
File: 2boys_looking_at_viewer_cowboy_s.jpg (477.1 KB)
477.1 KB JPG
>>108213122
>>
>>108213153
Easy. Green elephant.
>>108213157
Just don't care for it, simple as.
>>
>>108213155
I love novelAI. i convinced myself not to pay for it anymore because i have a 3090 and should be genning this stuff on my own hardware. but it takes a lot more effort to get the same quality and prompt adherence as novel AI.
>>
File: 1771784777843266.jpg (69.1 KB)
69.1 KB JPG
>>108213236
>we will make our own space
>WHY ARE YOU NOT OF INCLUDING ME BLOODY BITCH
fuck off
>>
File: 1752106014677080.png (456.5 KB)
456.5 KB PNG
In the age of information not caring about the current thing is the highest act of rebellion against the system.
This makes the shill mad, very mad, because not caring defeats them.
>>
File: 71c5e5ed774bc98b58c64fd05b0e5127.jpg (449.1 KB)
449.1 KB JPG
I guess it's too much to ask from /g/, but we should discuss books and academic papers more often me thinks.
Board quality would improve if people tried learning stuff for once instead of vibe coding slop.
We should have a SICP general, a TAOCP general, Concrete Math general, Intro to Algos general, and so on.
Or just a single general, to discuss whatever we're reading, like a tech book club. I will do my part:
https://files.catbox.moe/btjvow.pdf
https://files.catbox.moe/uyjmr9.pdf
>>
>>
>>
File: 1girl_tape_gag_bound_wrists_duct_tape_scared_evening_gown_silver_dress_st_s-54026998.jpg (260 KB)
260 KB JPG
NovelAI bros? we are eating good!
>>
>>
File: 88fbe73a-b5b1-447d-a95a-71e471efe583.png (1.7 MB)
1.7 MB PNG
>>108213122
Thanks for baking a NAI general!
>>
>>
>>
File: 1599921784361.jpg (82.9 KB)
82.9 KB JPG
>>108213320
>We should have a SICP general, a TAOCP general, Concrete Math general, Intro to Algos general, and so on. Or just a single general.
I'd post in it, I still haven't read TAOCP and Intro to Algos.
>>
>>
>>
>>
>>
>>
>>
>>
>>
File: 8e523450-8dfe-4929-8551-64f39972c2b3.png (1.1 MB)
1.1 MB PNG
I see
>>
>>
I like non-globohomo AI, or rather machine learning models for real things, backpropagation is a fun algorithm
Like if you want for example to estimate physical phenomenons, fitting their data to an approximation model, which can be non-neural too like polynomial and taylor series
>>
>>
>>108214543
I don't like the machine learning way of solving problems, it's more similar to statistical models than programming or symbolic models, it ends up being all about the data (which is why big data corporations love it) but incredibly opaque about the process.
Don't get me wrong it has its uses for certain problems, just like statistical algos have their uses, but I feel like it's been used for things it shouldn't and that it will backfire, there are already examples of ml ai tools failing catastrophically in medicine, and llms are another example of course.
>>
File: amazon.jpg (130.3 KB)
130.3 KB JPG
>>108213122
nice, ok so the futre is here and it's currently sucking, pic related, crashed amazon drone
>>
>>108214591
Sure, the data needs to be correctly interpreted
Differential equations are better for linear models, and non-linear models are too chaotic to be predicted
But in the right place is fun, not like LLMs but like, training a small dense model for assisting sensor fusion for example
Not AI as hype retardation but as part of statistical and numerical analysis
>>
>>108213122
>>108213131
Lain is an AI that manifested into reality. Fucking secondaries, the entire series is her realizing her true nature (guess what, that is why trannies latch on to her, they see themselves as finding their """true nature""" as women)
>>
>>108213811
Ah I see.
Well, if there's one thing I'll say to contribute, I hate how AI has been used to fuck up the web.
It's already been enshittified by SEOfags, but god-fucking-damn are there so many AI generated websites.
We need some fucking quality control, but major search engines don't give a shit (because they benefit from the ad revenue, of course).
>>
sry for being a massive retard but i read the op and still cant to get it right:
How do i attribute 2 different expressions to 2 different characters in 1 gen? Im trying the break prompt many t imes but it seem not to work.
>>
>>
>>
>>
>>
>>
>>
>>
File: Myarey_arafoo_otoko_no_isekai_tsuuhan_seikatsu_1.4__skindentation___no_text_s-2192247192.png (1.5 MB)
1.5 MB PNG
NovelAI general?