danicoro

[danicoro]

[danicoro]

Repeat after me: I draw so good. Not everything needs to be a banger. I'm not a content machine I'm a person who makes art and art takes time. Inspiration comes in waves and when it recedes that's when I should let myself rest

[danicoro]

Ellie and Chris are ready for a night out at a party, only Ellie quickly finds she has other plans when she meets Chris in the entryway.

OR

Two characters in an established relationship are insatiably horny about each other, and they will find any excuse to fuck like crazy.

[danicoro]

"Analytical AI is not the same as generativ-" Yeah, we know. AI of all kinds has been overhyped for years, and there isn't really "good AI" because of that. Algorithms and ideas that have their use-cases are far overpromised and overextended, their drawbacks swept under the rug, and public misunderstanding of them has constantly been fueled. On purpose. Before gen-AI, and before LLMs were used outside of translation, I was already complaining about the use of AI to control traffic systems, the use of AI in medical systems, the use of AI to replace parts of google's logic, the use of AI to judge job applicants, and other places. Long before generative AI, I was expounding about how analytical AI most often plays the role of garbage-data laundering, whether that's regurgitating racism to reject resumes, being used to generate evidence in trials via 'facial recognition' to secure convictions, (rather than to find guilty people), or covering up for bad data practices in astronomical studies.

(https://xkcd.com/1838/ , and this was in May of 2017. The explain xckd wiki page is also a time capsule of people's thoughts at the time: https://www.explainxkcd.com/wiki/index.php/1838:_Machine_Learning) (Here's another link from 2017: https://www.theguardian.com/inequality/2017/aug/08/rise-of-the-racist-robots-how-ai-is-learning-all-our-worst-impulses) Machine-learning can be a useful tool. I use ML algorithms. It has also been a tech-grift for literal fucking decades, and everywhere that the ML virus has spread in programming, algorithms have gotten worse, less accurate, slower, and much harder to understand. It's only been recently that we have developed the tools that finally put some of the internal linear algebra into easily parseable ideas, and we've seen just how much fucking worse it's been all along than even the cynics thought. It is highly likely that without the vast overpromising of what ML could do, that it would exist outside of academia, and I am not being hyperbolic.

Many of the more practical and established usecases now, (automation of production lines and quality control, for example), were previously programmed directly by programmers, who understood exactly what errors they might be introducing, and used simple ideas. The same tasks are now given to machine learning, require more time to develop, sometimes a lot more computing time, (as in, need to rent out a server farm for the training data amount of computing time), the errors need to be discovered, the error rates are not terribly better than before, and the programmers often have no idea how to properly fix the issue. A lot of times, they are nothing more than digital data rakers, spending all their time sorting out good data from bad. And because you don't know what the analytical AI is actually picking apart from the data, you also can't make the machines as multi-step or complex as before, as every step multiplies the error from previous steps. We haven't gained much from machine learning automation, which is where it should be shining, and we may well be doing worse than before. The simple truth is that the action of analytical AI has largely been to replace human labour with bad-to-mediocre automated results, often at a higher cost to the business than the human labour was, with an unknown error rate. Things people were previously fired for, have become the de jure standard. When it hasn't been replacing human labour, it has often been replacing working programs with . . . non-working or not-as-good programs. The places where it truly shines exist. But those are niche uses, and you're probably still overestimating the utility of them there.

[danicoro]

boutta get some sloppy toppy

[danicoro]

AI industry groups are urging an appeals court to block what they say is the largest copyright class action ever certified. They’ve warned that a single lawsuit raised by three authors over Anthropic’s AI training now threatens to “financially ruin” the entire AI industry if up to 7 million claimants end up joining the litigation and forcing a settlement.

well…darn

[danicoro]

being a writer is constantly google the definitions of words you already know the meanings of because your brain's always paranoid and telling you maybe you've been using them wrong your entire life

I can excuse misusing words in my daily life but my mlm slow-burn enemies to lovers smut has to be perfect

[danicoro]

[danicoro]

You can survive almost anything through the right combination of:

Bitching and moaning

Hater-ology

Doing a goofy little bit about it

Having a buddy say "that's so fucked up" at intermittent points (you can also be your own buddy)

Destroying the cursed amulet you carry everywhere, why do you even have that thing

[danicoro]

WHEN ON PERIOD:

do not crash out

your feelings are NOT valid

do not send that text

don't kill yourself. lock in

do not act on negative emotions until at least 2 days have elapsed

[danicoro]

and with your help it can rack up 700k notes on tumblr in 2024

no tumblr this doesnt need tags im releasing it into the wild as god intended

[danicoro]

Bitches will play a whole ass game for like 3 cutscenes with a character they love just to feel something

[danicoro]

''what if my writing isn't good eno--'' what if it's a reflection of your soul. what if it has a place in this world. what if you write it anyway

[danicoro]

This is a picture for those who think being a writer is easy: WE FEELING STRESSED ALL THE TIME!!!!!

[danicoro]

“protect children” <- reactionary drivel basically every time

“be kind to children” <-radical thinking that causes way more arguments than you would ever imagine

[danicoro]

https://twitter.com/delaneykingrox/status/1090402436995473408

[danicoro]

“is this character good or bad” “is this ship unproblematic or not” “is this arc deserving of redemption or not” girl…