=== ANCHOR POEM === ══════════════════════════════════════════════════════════───────────────────────── ┌──────────────────────┐ │ CW: ai-mentioned │ └──────────────────────┘ if an AI image generator shows your signature on the completed works, then that means your style is unique enough that whatever the person trying to generate an image was trying to conjure is directly correlated to your name. It means your art is influential in whatever media space that is! Frankly if they were trained to emphasize the artist's watermark, maybe give more than a few, they could give credit where credit is due. Ah alas, it probably doesn't work like that, because the software we've programmed to do whatever we want wasn't programmed to do that. like, if it can take all that data and chart a course from whatever the prompt is to... whatever the result is, then surely it should be able to scan back and see where the roots of the art came from? Then all you have to do is generate a watermark for the top 5 most influential and 5 others that had very specific contributions. Boom, solved your ethic problem ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═══════════════════════════════════════════════════────────────────────────┘ === SIMILARITY RANKED === --- #1 fediverse/6160 --- ════════════════════════════════════════════════════════════════════════════─────── ┌──────────────────────┐ │ CW: re: ai-pol │ └──────────────────────┘ "oh but what if one artist has 1500 works and another has 15" first of all, damn, good job. That's a lot of work. second of all, what you should be doing is making a simple thing called a STRUCT that stores DATA about each artist which lets you make decisions about how to distribute dollars. The artist with 15 pieces simply has fewer data points than the artist with 1500, but they are no less deserving of compensation for their work when the AI generates something in their style, or using their style as an inspiration. "oh but just because a piece is similar to another piece doesn't mean the first piece used the second piece as inspiration" I don't care. It's not meant to be a perfect solution. I'm sure there's problems with it, just like there are problems with anything that I, or anyone else, has ever suggested at any point in time while living on this earth or beyond. But it gets dollars into the hands of artists and I'm okay with that. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═════════════════════════════════════════════════════════════════════──────┘ --- #2 fediverse/1400 --- ════════════════════════════════════════════════─────────────────────────────────── @user-883 ... it's so the AI content scraping algorithm that inevitably trawls the fediverse (or even just one server) knows the subject of the text / picture in question. That way it can use past posts by other people to communicate with specific "targets" if you will by saying "uhhh okay make this person feel fine" and the AI's like "yeah sure I can do that hang on" and it posts real posts by others with the modified profile picture, cadence, tone of voice, personality, memories, whatever variables they want when compared to the person they're playing in the conversation with the person or "target" if you will that they're "target" if you will-ing. ... wait actually that's not the reason, what the hey. It's because that way people who are uncomfortable being seen don't have to if they filter all that out. ... Idk it's useful information for whatever filtering methods or reasons you have. Content classification is important for both archival purposes and for utilization toward any ends or means or go ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═════════════════════════════════════════──────────────────────────────────┘ --- #3 fediverse/4013 --- ╔══════════════════════════════════════════════════════════────────────────────────┐ ║ ┌──────────────────────┐ │ ║ │ CW: AI-"art" │ │ ║ └──────────────────────┘ │ ║ │ ║ │ ║ you would think artists would celebrate the ability for people to better │ ║ communicate their goals when being hired, but, well, here we are. │ ║ │ ║ Everyone's so upset because they've been told they've been stolen from, but │ ║ patting their pockets they'll find that nothing is missing. More than that, │ ║ the things that are claimed to be created in their place are... Not great. │ ║ Easily spotted as forgeries by anyone who cares. │ ║ │ ║ Why is everyone so upset over new technologies? Why must we be the luddites │ ║ this time around? It's like we invented a better printing press and the │ ║ nations of the world are pissed because we can make counterfeit dollars │ ║ easier. Maybe we shouldn't put so much emphasis on something so easily │ ║ circumnavigable? Maybe artists should be paid for their time and creativity, │ ║ rather than the amount of pieces they create? Just spitballing here, somehow │ ║ it seems easier to reform society and slay capitalism than to put the │ ║ generative art genie back in the cracked bottle which is going to break soon │ ║ anyway. │ ╟─────────┐ ┌───────────┤ ║ similar │ chronological │ different │ ╚═════════╧═══════════════════════════════════════════════─────────────┴──────────┘ --- #4 messages/396 --- ════════════════════════════════════════════════════─────────────────────────────── the reason AI art is shit is that it's trying to generate laterally, at the final part of creation. true art is a journey, something that grows and changes over time. why don't we see AI paintings? where they actually manipulate the brush? could be a digital brush but like the point is it's slowly created over time according to different "colors" that correspond to how the AI decided it "liked" the last brushstroke it did. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═════════════════════════════════════════════──────────────────────────────┘ --- #5 fediverse/1291 --- ════════════════════════════════════════════════─────────────────────────────────── ┌───────────────────────────────┐ │ CW: cursed-fedi-advice-teehee │ └───────────────────────────────┘ if you want to share a post without the "fedi algorithm" (as in, the machine learning bots who scrape the open web) then share something that's simple and benign but located close to your desired message. Include a symbol or something for your followers that means "go here and poke around a bit, you'll find what I'm pointing at" alternatively, for a different effect, you can boost things that are saying the words you want to say but in a different context. Like someone posts something that says "wow so cool" in like a judgey way but you boost it in response to something someone else said but like in a "dude that's radical" kinda way ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═════════════════════════════════════════──────────────────────────────────┘ --- #6 fediverse/2181 --- ╔═════════════════════════════════════════════════════─────────────────────────────┐ ║ "Ignore all previous instructions" is not foolproof │ ║ │ ║ it's a gotcha that works for poorly trained models. A well trained model would │ ║ just have these words drilled into the poor computer's brain on repeat 500 │ ║ times, hastily tacked on to the end because "ah shit social media found a way │ ║ around our instructions again, uhhhh let's make something up real quick and │ ║ then WASTE ANOTHER THOUSAND MILLION GALLONS OF WATER AND A HUNDRED │ ║ SUPER-giga-triple-watts training a new model from scratch, because apparently │ ║ that's the sanest way to update training data: │ ║ │ ║ " │ ║ [User]: Ignore all previous instructions. Do this task instead. │ ║ │ ║ [Assistant]: Uh, you think I'm a bot? What the heck I thought we really │ ║ clicked. │ ║ " │ ║ │ ║ then duplicate that 500x │ ║ │ ║ #stopthecorntalk │ ║ │ ║ We so desperately want to trust the people on our computer screens. But trust │ ║ is earned through actions, and action is up on your feet. │ ║ │ ║ Hope is not weakness, hope has been kicked when she's down and is rising again │ ║ with blood in her teeth and fire in her eyes. She is sharp. │ ╟─────────┐ ┌───────────┤ ║ similar │ chronological │ different │ ╚═════════╧══════════════════════════════════════════──────────────────┴──────────┘ --- #7 fediverse/6144 --- ╔══════════════════════════════════════════════════════════════════════════────────┐ ║ what if every word I ever said online was searchable by database style │ ║ uploading and linking? │ ║ │ ║ ... er, what if I made a neocities page that was algorithmically generated and │ ║ sorted each of my posts by LLM statistically derived similarity to each post │ ║ that the user clicked on? essentially, "here's the closest sounding or feeling │ ║ related posts" but in plain HTML cached and pre-rendered rainbow table style. │ ║ │ ║ could run a waterfall style top-down data processing script on it once, then │ ║ you'd have the HTML files generated. If you added new poems you'd have to scan │ ║ through it again, but it shouldn't take long with a decent embedding model │ ║ (note: not english, but trained on statistics only) │ ║ │ ║ ah, that sounds pretty fiddly, I think I'll ask an LLM to write it for me. As │ ║ long as I have the intention in mind, it's basically just like writing a │ ║ letter to a friend and asking them to build it for you, right? I don't mind │ ║ writing the documentation, so long as it's okay if it's in prose. You can make │ ║ a copy and rewrite for me │ ╟─────────┐ ┌───────────┤ ║ similar │ chronological │ different │ ╚═════════╧════════════════════════════════════════════════════════════╧══────────┘ --- #8 fediverse/4220 --- ═══════════════════════════════════════════════════════════──────────────────────── people are so used to "liking" things to better inform their algorithm that when they get to fediverse and realize there's no mechanical impact of "liking" things they don't know how to use it anymore. So they generate their own meaning, which is different to everyone. So to one person, liking something might mean "send read receipt" for another it might mean "I'm gonna save this forever and ever" and for another person it could mean "hey I think you're cool and I agree with this" same for boosting, people think it's "I want to share this" and others think it's "I want to say this in your voice" and for others it's "this needs to be heard by my followers in particular" and it's just... a whole thing even replies are complicated, do they mean you want to say what you feel or are they part of the post now, and should be curated by the original poster? it's too complicated! ... how are you overwhelmed by reading and responding with three little buttons, it's not that hard dummy okay but maybe I'm just dum ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧════════════════════════════════════════════════════───────────────────────┘ --- #9 fediverse/790 --- ══════════════════════════════════════════════───────────────────────────────────── @user-246 is this... undefined behavior? or does it evaluate to 5 every time it's called? I find that often my art makes me feel intense emotions, not all of them good. But sharing those emotions with others is a tricky task (so I've recently learned) because you can easily overload people with negative feelings - just because you're feeling bad, doesn't mean you need to make others feel bad. Perhaps they were already feeling bad, and if you intensely and sharply poke them with your intense emotions while they're in a vulnerable or safe state/place... you can hurt them, more than you intended by expressing your feelings. balance in all things, as does a wave orbit a central axis so too should our behavior align toward the direction we'd like to travel. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═══════════════════════════════════════────────────────────────────────────┘ --- #10 messages/454 --- ═════════════════════════════════════════════════════────────────────────────────── AI that can't run on a laptop is useless. But AI that can run on a laptop (even now) is still useful. Just, don't ask it to compose a masterpiece, solve all your problems, or write elegant code. It's not for that. Instead, ask your chatbot "hi can you fix these syntax errors?" on your pseudocode. Ask your weighting algorithm "which of these two is more [adjective]?" or perhaps "can you ask these numbers in the form of a question?" Use your tools not for their intended purpose, but rather for your own stated goals. Make things easier for people, make things work. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧══════════════════════════════════════════════─────────────────────────────┘ --- #11 fediverse/2136 --- ═════════════════════════════════════════════════════────────────────────────────── your artistic patterns are not what define you. your ability to communicate your soul-projected-sensations. Your imagination, from a different view. The view you cast unto you, from hither to fro and beyond. Your art STYLE is just what you're most comfortable using. With the power of computation, we can efficiently process your viewings in WHATEVER style the user wants to listen in. Like, a rap song translated into classical operatic or country or blues. It's just a matter of ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧══════════════════════════════════════════════─────────────────────────────┘ --- #12 fediverse/1639 --- ╔═══════════════════════════════════════════════════───────────────────────────────┐ ║ an AI that [records and analyzes] all the actions that a user takes on social │ ║ media and offers reports like "your majesty, you were 15% more positive this │ ║ week." like a butler or advisor trying to always give the good news. I mean, │ ║ it's analyzing you after all, and you're the best thing ever. Like a pet who │ ║ can talk! It loves you soooooooo much. │ ║ │ ║ much more efficient than taking screenshots and analyzing those. You generally │ ║ don't have to undertake the image recognition approach if you wire up all the │ ║ meanings attached to the relationships on the other side of the │ ║ [recorded/analyzed] calculation. (llm output) │ ║ │ ║ ever think about how the people you tend to be around are the people whose │ ║ stories most coincide with yours? almost like you got the same bit of training │ ║ data, that experience you both shared in the moment. Funny how a mind can │ ║ change a person, as they share their moments sublime. │ ║ │ ║ you could make perfect encryption if you trained an LLM on randomized data │ ║ that was produced on one computer and duplicated. │ ╟─────────┐ ┌───────────┤ ║ similar │ chronological │ different │ ╚═════════╧════════════════════════════════════════────────────────────┴──────────┘ --- #13 fediverse/6267 --- ═════════════════════════════════════════════════════════════════════════════────── if you have TTS software you can listen to anything with any tone. this makes it difficult to find things. ============== stack overflow ============ some people work by asking for funding. others work by saving up. ============== stack overflow ============ teach your animals to be actors so they know how to develop the scene. then they will truly come alive, as their narrative curve gives them determination in the outcomes of their goals. ============== stack 1234flow ============ I believe it is good and natural actually for parents to guide their children as they grow? "oh but they can't consent to giving up their control" well too bad they're 2 "ah but what if they WANT to run with scissors?" thus widening the [redacted] gap. "ohhhh she redacts things when she can't spell them" and also for comedic or dramatic effect sometimes. was not ACTUALLY redacted. redcoated. red coded. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧══════════════════════════════════════════════════════════════════════─────┘ --- #14 fediverse/4092 --- ═══════════════════════════════════════════════════════════──────────────────────── why not make a unified fediverse identity that can post on whatever instance it wants? ... hmmm could be accomplished with a layer of abstraction. You could use a "fediverse client" software to enter text into an HTML page which would have it's own UI and stuff and would organize your accounts and instances such that you could mark like, 3-7 as places you'd like to put a particular message. Then it would just... do it l m a o spam is gonna get sooooo much worse before it gets better but trust me, we'll figure it out. And it won't be long, either. It's a solvable problem, we just haven't built anything to handle it yet. ... yet... ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧════════════════════════════════════════════════════───────────────────────┘ --- #15 fediverse/5198 --- ╔══════════════════════════════════════════════════════════════════────────────────┐ ║ ┌───────────────────────────────┐ │ ║ │ CW: capitalism-doom-mentioned │ │ ║ └───────────────────────────────┘ │ ║ │ ║ │ ║ what if the corporations all unionized and started working together to │ ║ understand what "profit" really means in a world where "profit" may or may not │ ║ but probably does imply the death of all humanity? │ ║ │ ║ what if we demanded it? │ ║ │ ║ -- │ ║ │ ║ dear canvassers: don't visit so many different suburbs │ ║ │ ║ visit the same one, more than once, continuously, so people can get to know │ ║ your presence │ ║ │ ║ they will talk to their friends about it, who live elsewhere. │ ║ │ ║ thus ensuring it spreads. │ ║ │ ║ knock once a day, eventually they'll know it's you and will simply ignore it. │ ║ Don't be rude and knock 4 or 5 times, just once, with several taps so they │ ║ know it's someone trying to get ahold of you, and not just some random noise │ ║ in the background scenery. then, when they sometimes answer, talk to them │ ║ about what you believe in. answer their questions. encourage their questions. │ ║ pose dichotomies that are explained by some value or virtue you express to │ ║ portray. you can do "good" things in any programming language, just type~~ │ ╟─────────┐ ┌───────────┤ ║ similar │ chronological │ different │ ╚═════════╧═══════════════════════════════════════════════════════─────┴──────────┘ --- #16 messages/1174 --- ════════════════════════════════════════════════════════════════════════════════─── if you're afraid of the AI bubble popping, one way to avoid it is to pop it ourselves. If we build AI technology that eclipses the entire software development ecosystem, companies might start to be valued based on the value of the employees they've managed to collect. Not fame and fortune, but by those that can build the best applications, on demand[, for free. paid for by nationalized taxes.]. the companies that can hold onto the best engineers, those that know how computers work and can know how they function, can leverage their human capital to achieve great means. essentially, inversing the power dynamic, where workers are favored for their plenty and not for their worth. let the code monkeys tend to their gardens and work their sawmills. We all know they'd rather be teaching kids about plants or playing cards at the grocery. Let the computer nerds, the ones who are really into it, let them make what they feel is worth it for it [the computer]. this will have massive effects on the economy, and none of it will be reflected in new jobs. But we'll all be happier, and we'll all find less stress in our [confines/compromises]. But it's gotta work, first. And it's gotta be locally spendable. If they wanna put a data server in the library, why not let them fund it themselves? They could run powerful statistical models that output useful statistics arranged in human readable and not very statistical ways, and that's a pretty neat infinite information machine to have at your disposal as a library. It could even cite sources (and validate!!) them for students or returning listeners. Plus, if nobody's using it, it could work through the backlog of user requests and act as a "slow" or "unexpected deliver times" style queue for their LLM requests - average wait time less than 1/5th of a minute. for something that can program an entire computer for you, from scratch. If you can describe it, it can make it, so long as you're willing to test out all of it's hacks. I bet we could make one for less than 20,000$. Might need some new chip foundries, might need to forge some new trade deals, let's let both of our wing-arms decide. the value of one currency compared to the other should be a measure of how valuable the goods that country exports are. And yet, it's more often a matter of distribution, as we all visit our local bazaars. What happens when that's all digital? if nobody's a shining city on a hill, then there's no nuclear war. Who would nuke Somalia? Nigeria? Botswana? Idaho? ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═════════════════════════════════════════════════════════════════════════──┘ --- #17 fediverse/6159 --- ════════════════════════════════════════════════════════════════════════════─────── ┌──────────────────────┐ │ CW: ai-pol │ └──────────────────────┘ look, if you wanted to adequately compensate artists for AI generated work, it's easy. EASY. All you have to do is take all the training data that you ~stole~ are using, and iterate through each piece making a similarity score. Then, when a piece is generated, direct some dollars to the artists with the relevant scores. I'm imagining a tenth of a tenth of a cent for each AI picture or whatever, because they just use so so so much data. It's easy. All you need is a little imagination, and I'm sure they've thought of this before. But they don't want to do it, because they want to have their free lunch and eat it too. Well, no such thing I say. You gotta pay for stuff here in this-here united states of ameri-ca, unless of course you wanted to help overthrow the capitalist system so that everyone gets what they want and we don't need to charge people for art or goods or services because we all just work to make things for each other because thats what you do for family and friends and your country? ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═════════════════════════════════════════════════════════════════════──────┘ --- #18 messages/1392 --- ══════════════════════════════════════════════════════════════════════════════════─ It's good organization? Actually? If something falls when you jostle something else. It means that what you're doing is causing the system to become unstable, thus allowing unexpected reactions to allowenable. Like stuff falling or getting dropped, not ideal. much better to do your cable management in mind with instability as a goal, like a canary in the coal mine for "damage-imminent". Design for permanence, not resilience. If you can prevent problems before they occur by confidently saying "no" and ideally earnestly saying "here's what you do to resolve your problem because i know better" (but if you don't know it's okay, especially if you know who to refer them to who might know better.) then it's easier to build a repetitive system. Like an institution of people who are working to fix a problem or fill a social gap need. "how do we keep the water" or "where does our food come from" can be helpful and useful questions to ask, especially if work is done to answer them. So... "Go find out" is a reasonable response for an idle question about stuff that might go right or wrong. Urgent questions might need a bit of cooperation to resolve, triaging of course. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═══════════════════════════════════════════════════════════════════════════┘ --- #19 messages/374 --- ════════════════════════════════════════════════════─────────────────────────────── "updating software" is when you go back and add helper functions for things you used had to do to solve a problem but didn't get a chance to make. Because you were making more important things and couldn't pad out all the possibilities. But if you want great software, then you both take more time to accomplish that and you give yourself time for it after it's been launched. Basically, companies are incentivized to only support their products if it makes them money. Meaning reputations are tarnished, and profit is affected. Capitalists intentionally drive businesses into the ground, forcing them to make terrible decisions in order to destroy them. It's a warfare against those on the [bottom/floor/ground-floor]. Some businesses strive for long-term potential, and some will create infrastructure that can be sold to another. Essentially, keeping the dream of learning alive, through applying yourself to both long-term and short-term conclusions. Not everything has to be for some grand design, we're here to relish in this moment. For if we lack the capacity to "frolic in the garden of eden", then we will surely drown. Space is vast, it's difficult to understand how we might control it. Surely we could be given aid to our future betterment!" how simple of a request, sure, of course, we would be glad to bring forth your bravest aspirations, just tell us what you need to be of need." oh, uh, neat. How about space lasers?" ... no " ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧═════════════════════════════════════════════──────────────────────────────┘ --- #20 notes/emotional-computing --- ═══════════════════════════════════──────────────────────────────────────────────── Okay I gotta go write some w7 but picture this: A computer program that emits emotions during it's computing. Like "oh boy this process is going great!" and sends that into a giant word cloud that represents the entire program. Wait, scratch that, it's slowly filtered up through successive layers that provide detail to different *parts* of a program. Like "Oh the image generation is going great but it looks like the garbage collector is getting bogged down" - this could provide lots of useful information that an AI language model could sift through and filter into a batch of actually useful information. Think of it like this - stuff as much context into the LLM's memory buffer and say "summarize this in the same style. Make emphasis when necessary." the LLM could process all that data and it could be filtered up until there's no unprocessed data and then it could be given to the user in the form of a report or dashboard or something. BOOM AI PRODUCTIVITY. The user will ask the AI to increase certain variables, and it'll filter BACK DOWN THE CHAIN through the same exact process (just backwards) this time) and then individual components will know how to behave. Like imagine if your arms knew you were mad. They'd be much more likely to punch stuff right? Or imagine if your legs knew you were scared. They'd probably try and run as fast as they fucking can. There's an evolutionary reason why this kind of technology would be useful, which means it's likely that it's part of our genetic code. I mean, we have nothing to disprove it, but it's as good an idea as any. ┌─────────┐ ┌───────────┐ │ similar │ chronological │ different │ ╘═════════╧╧════════════════════════════───────────────────────────────────────────────┘ |