Artificial intelligence told me how to spin yarn. I have thoughts...
HTML-код
- Опубликовано: 1 окт 2024
- The potential impact of this technology on niche communities like fiber art is increadibly concerning and I need to express my thoughts about this. There is a shift happening in online content creation right now that feels like we are moving from a spinning wheel to a spinning jenny. This is a whole episode of deep thoughts with Evie and how AI has the potential to change fiber art and the ways we talk about textile history in massive ways so buckle up. WE NEED TO TALK.
Please support human artists!
🛍️ JillianEve Shop - jillianeve-fib...
❤ Support The JillianEve Channel by leaving a tip for Evie through Ko-Fi - ko-fi.com/evie...
Want more spinning content? Check out the JillianEve Patreon to join the Discord and more! - / jillianeve
📧 For business inquires admin@jillianeve.com
🎵 Music used in JillianEve videos is licensed through Artlist.io
__________________________________________________
~~ Social Media ~~
📷 Instagram / jillianevefa
🎉 Patreon / jillianeve
📰 Blog www.jillianeve...
⏰ TikTok www.tiktok.com...
_________________________________________________
*I use some affiliate links. They will always be clearly marked and you are never obligated to use them. If you make a purchase through an affiliate link, I may receive a small commission at no extra cost to you. Thanks for your support!
__________________________________________________
As an artist with very severe imposter syndrom (I can't even upload anything yet), more and more artists are starting to give up on the fight of artist vs AI. PLEASE do not give up the fight. This can be a very beneficial tool of used correctly, though, many people will not be using AI as a beneficial way. It has to be controlled before it is too late.
The saddest thing is that we (my friend's crew) were expecting the day when robots will do all the boring jobs so the human race can benefit from them for daily tasks, and to be left to relax and create art. But now instead of my vacuum cleaner to give me more time to sit and write crochet patterns, it will generate them in a few seconds, and probably will start to sell them!!! (I'm creating amigurumi patterns only for my personal joy, but I know the vacuum dude pretty well, it will want to sell!)
_That's_ how I feel about it. If you're not making money off the piece and you're just using it to get an idea of something, *and if it's trained on work from consenting artists* I don't really care
What I dislike the most about the current use of AI technology is that it has the power to free People from a lot of menial tasks, but instead it is directly used to make art and creative content - stuff that gives people joy and makes us unique.
But I loved your weird and creative take on how to present this video, even when it was inspired by AI art!
Exactly! Robots are supposed to be doing the manual labor while we do the creative labor. Then there's that robot that was given a repetitive task and decided to turn itself off instead... hmm.
@@sarabarnes7189 if I were that robot: honestly same 😅
Exactly. As with most things, the problem isn't necessarily the technology, but who's controlling it (or choosing not to, or "teaching" it on property they didn't have the right to), and for what ends ("making more money for the sake of making money" rarely results in the betterment of humanity; we don't need to posit more nefarious ends than that to explain some truly horrific means).
omg I saw the hottest take on Euronews about "what menial tasks can AI free you from doing" and it listed writing and graphic design. Then the article concluded by saying all these 'menial' tasks being freed up means you now have more time to be creative. Like, hello? The literacy and awareness was just not there
@@bilunala insert .gif of very fast blinking dude here
That was a great conversation Evie. and hit on a lot of my fears but also dug way way deeper into aspects I hadn't even considered. As someone who loooooves history and learning new things..... I could see myself digging into AI, but then it not citing its sources...and it not giving turely accurate information.....while it's not there yet It may be there very quickly. And the bit about consumerism (paying for their product to be mentioned) and bias especially racially in many of those images.... the implication is HUGE. and quite terrifying. I have a lot of thinking to do on this. I'm excited about the aspect of "pooled knowledge" but also know how quickly that can become manipulated......Thank you for bringing this video. I think it's something that really needs to be discussed a lot more about a lot of things. It's a fun now but has major implications for how things are used and taught In the future.
I find the concept of AI terrifying. Thank you for broaching this thought provoking topic.
Thanks for doing this! I shared it in a Community post on my channel. I think we should be more conscientious of this and some of the dangers of AI.
Thanks!
I’m à French follower, I support your work and absolutely love your work even it’s the first time I post a comment on one of your video. Amazing, interesting, concerning…
Thank you, Evie, for this educational (although alarming) presentation of how AI works. I appreciate your focus on the fiber community, but I see the ramifications of this spreading far beyond and into many other areas of our lives. Thank you for raising our awareness of this problem.
This might be my favorite video of yours.
10+ years ago, when blogging was a more major thing, I remember the difference between the blogs that were from and about people working to really share their experiences and thoughts (some of which were monetized and some weren't), and those that seemingly existed *only* to sell... something. Ad space or sponsored products or what have you. And I remember having some of my work (the former) be repackaged and sold to a conglomerate of the latter style, and how angry I was, and how hurt, and it absolutely contributed to me walking away from my writing, for a very long time. I remember how important it was AS a writer to reference back to any other writer whose work and thoughts influenced my own, and how outraged we were at the writers obviously responding to a community conversation WITHOUT linking back to it -- ie, stealing our thoughts and our ideas and our work without citation.
And now it's happening to artists and creators of all stripes, wholesale, and seemingly without recourse. And without any individual humans actually responsible for it.
I'm not a technophobe, by any means. (Sorry, you're never going to outthink Commander Data!) But I absolutely agree that it's crucial that we take steps to product ourselves and each other -- and future newcomers to our crafts -- from the theft, the bigotry, and the misinformation that is clearly happening and will continue to happen if these technologies aren't redirected and guided by actual, ethical, human beings.
100%! I wish I could like your comment twice :)
Thank you for sharing your experience. I'm so sorry that happened to you!
The problem I saw with those instructions (I am a very new spinner) is that although nothing was actually inaccurate, it generated more questions than it answered. For instance it talks about drafting the fibre, pull off a small amount of fibre but which hand,which fingers, how do you hold it all. When I was first learning I relied very much on watching what you and other spinners were doing with your hands and copying your actions. Words can't tell you that. Although experienced spinners and even me who is a beginner can understand what is being said and follow it, when I was very first learning, this text would have just confused me. It would have been pretty useless.
And then it is pretty scary to think of how AI could be used by unscrupulous and/or greedy people. Thanks for this video. Much food for thought.
I must admit, I was thinking "What would AI have to do with fibre arts?!". But you make a great point about how all nuance and diversity could be lost when AI just echoes stuff it finds on the internet with zero judgement. I also find it really odd that most AI bots can't seem to show their work, aka their sources.
Until yesterday I had given very little thought to AI technology. I presumed it to be a not-very-accurate-curiosity, not good for much of anything except amusement -crazy knitting patterns and horrible recipes, not very good art, etc... A friend had posted a YT video to her Twitter feed about an interesting video she watched. I watched it, too, and became increasingly alarmed and convinced as the subject of the interview had incredibly valid points and cogent arguments against "open source" AI and it's impact on our species, globally. It cannot become benign without human intervention creating "alignment" within the system. And no one has determined a way to do that, yet. We are quickly running out of time to walk this horse back into the barn. Whether or not you believe this gentleman, he seems to be one of the smartest people on the planet, and I think we should listen. Even the interviwer had trouble grasping some of the points, but I was able to understand the fundamental concerns on a human level, if not on a technological one. For those interested, it was The Logan Bartlett Show, episode 63, "Eliezer Yudkowsky on if Humanity can Survive AI". At over 3 hours long, the information is alot to digest and internalize, but it's so very important to consider. And, yes, Evie, the AI stating that Merino is difficult to draft for beginners seems like it did use you as an uncredited source. Thank you for this video, and for creating a safe place to discuss this. Hugs, always.
Thanks for the show/interview info. I'm heading there next. If it's too deeply technical it will probably take me sever bites to get through it so it is digestible.😉
This is scary… it will affect everything we do and most parts of our lives whether we like it or not 😢
I don’t disagree with your concerns at all. Honestly, though, these same problems exist with random people on social media. So many people use Andean, Navajo and chain plying incorrectly. It has been a problem since social media started. The AI is calling it the wrong thing because PEOPLE have been doing it for a long time. There has been no accountability to people online, by and large, for too long. So often people repeat something they heard someone else say on some site with no experience of their own. It happens constantly in the fiber community.
As mentioned, I don’t disagree with your concerns. It’s scary how people will believe anything they see on RUclips or any other number of sites with no fact checking whatsoever. Now it extends to AI. This is a problem that has existed in some form since people have existed. The checks and balances seem to be disappearing.
Very interesting. How can an AI be programmed for critical thinking, to determine what is true, when humans do not seem to be capable of it.
Lol… exactly!
In that case then, it’s hoped AI will find the error and correct it 🤣
I’m so glad you tackled this topic. You addressed so many aspects of AI and it’s possibilities as well as shortcomings. Your examples illustrated your points clearly and with context to our craft. Thanks!
Your description of time spent on the internet, items arriving days later, tens of windows open on the laptop and hours missing sums me up perfectly. Am I neurodivergent?
Technology isn’t always neutral. Especially under capitalism
I saw this on Twitter but someone said "Me working the underpaid labor jobs & robots making my art wasn't what I wanted but here we are..."
Wonderful video. Thank you. I'm not a spinner, but was refered to your video by a friend for the excellent points it makes about AI use and the problems it will generate if we don't address them.
NOTICE: This comment was generated and edited without any AI tools used.
IDEA: I may start adding the above text/statement to any work I do. The idea being that it could help the reader/viewer in judging how much stock to put in the ideas set forth. 🙂
Wow! That's a deep topic. Thank you for opening up the conversation.
As someone who can't get the hang of 20thC tech, I'm struggling with the recent changes in AI. There's going to be a lot of adjusting over the next few years.
I suspect in the end, it's about trust.
When I learned to spin, I read books. And magazines. And by then there were websites. Some of these were written by people who knew what they were doing, and others were regurgitated by people eager to make content to fill a quota. The latter caused a lot more harm than good and I learned to find the chain of trust to understand which voices to listen to more. In the end, I spent two years unable to spin even though I knew all the details of how it should work. It wasn't until I took a class from a local spinner that I could make yarn.
With AI we have a change in how we view the chain of trust. I suspect this will mean more for society than any other technical change of the last 90 years.
This is likely to result in an increase of in personal learning like classes at fibre festivals or through local guilds and other communities. Or should I say "the desire for in person learning" because we're quickly losing that option for so many people. More and more I feel that we need to keep these in person options. Be it a counter-culture fight against the machine, or simply because some people are like me and learn better by doing than by reading.
I totally agree - we need to keep the in person options. But in order to do so, we need to have enough people who are still interested in learning what we have to offer.
Especially for many younger generations, it is awfully tempting to go the seemingly easy and faster way. As soon as it gets tedious, they drop off…
Great conversation. It does give me a heavy heart because of the damage that AI is/will cause.
I first encountered websites claiming to let you use "AI" to write blog posts and articles over a decade ago - for a price, of course. This technology isn't new. It's just become more widespread, free to access, and yes, more sophisticated. ChatGPT moving to an open beta (so the creators can refine their model with the unpaid labour of its beta testers, I might add) has shifted the conversation around what algorithms are capable of, for sure.
Though I wonder what's going to happen when the team behind it inevitably start asking for money to continue using their shiny new toy? Making a cool new thing both free and widely available is just the first step to demanding money from users - we've seen it happen with Facebook, Twitter, and other social media platforms. It's only a matter of time before these new technologies go the same way. (Corey Doctorow calls it the enshittification cycle, which I think is apt, since user experience tends to get worse and worse as time goes on.)
Also a technical note because I am That Kind of Person - ChatGPT and Dall-E etc are not AI. None of it is. There's no Intelligence in there, artificial or otherwise. AI is just the trendy buzzword being used to describe it. ChatGPT is a Large Language Model, which means it's analysed lots of words and how often they appear next to each other, then attempts to replicate that when asked a question. (It really likes starting its concluding paragraphs with "Therefore" or "in conclusion", have you noticed?).
While we talk about AI, it's important to remember that what were talking about here, really isn't AI. AI is artificial Intelligence... ie: a machine that can think. These modern "AI" programs mimic AI, but what they're actually doing is the same as the great YT algorithm has been doing for years. They look for certain keywords, and search that information, then put it together. They're usually set up with basic grammar and syntax programs, and what they do is evaluate what it's parameters deem "relevant", then regurgitate that information. The trouble is, it's NOT intelligent..... the only way it can tell if something is relevant is how many times it's repeated. As we know, this is how misinformation becomes gospel for many people. There's a great danger in handing over information curation to a programme, which is that it has NO IDEA what it's presenting. The only critical facility that an AI programme has, is the ones built into it. Yes some of these DO "learn" but as the old adage goes, garbage in/garbage out.
As a person with an education in computer science I find that you describe the problem fairly well, but are still contributing to the misunderstanding. Why are people always focusing on the technology instead of the actual problem? The problem here is misrepresentation and the spreading of misinformation. A very old and an entirely human problem, something AI can only mirror or as you said, amplify. It essentialy enabling us do it faster. Something which was also true of computers, of TV, of radio. Of books even. Social problems require social solutions, and by focusing on the technology people are ignoring the root cause.
My main problem with chatGPT is that it should be required to provide a list of its sources, as should any article. Fact checking is, as always, up to us.
Very good thinking, as always. Thank you!
I just HAVE TO comment, because something parallel gave me such a laugh: The first time I was playing this video, at 16 minutes and something, at "That's it, we solved the problem. Don't go there and we'll be fine!" my graphics card overheated and my ancient laptop shut itself down. Not in the middle of the sentence, but at the exact end! As if that really were all there is to it.
As humorous as this was, it got me thinking. One of the problems that AI is compounding is our short (and shortening) attention spans. We have created a technology culture that does not discern between complex background processes and motivations for our actions. The machines analyze our behaviour, not our minds. Idle pastimes are dumped together with all levels of interest, even serious scholarly research. The value of content is measured by the clicks it gets, not the engagement of minds and souls. The worth of an article is determined by the attention it gets, not how well it answers the question it was written for. Or what that question was. Or if there even was a specific question. Everything is lumped together, and whatever floats to the top is touted as so obvious that "everyone knows" (and no credit needs to be given).
We cannot blame only technology for this crop we are now reaping, just because it has cultivated the field. The seeds were already within us. We have evolved to make use of the sparse energy-rich foods when they are available. Finding the rare delights encourages us to look for more, in order to supply us on the long haul where there is less excitement. But those resources are no longer scarce, and the long haul is seen as undesirable. We have become children who only ate the chocolate bits from their cereal, and now throw a tantrum if not all the cereal has chocolate in it. Never mind who got enslaved to produce that chocolate.
We are also internally wired to look for patterns and apply explanations based on our own minds to phenomena outside ourselves. We, as a species if not as individuals, are inherently biased toward believing that another mind created an event, such as an AI consciousness rationing information by shutting down a machine at a certain time. And this bias is also fed into the artificial intelligence. Are we perhaps becoming a cargo cult?
en.wikipedia.org/wiki/Cargo_cult
I think you're addressing very specifically the misgivings that so many of us have difficulty articulating. Particularly problematic is the lack of source citation. AI that generates content so quickly should be able to leave a massive bibliography in its wake. It's only inconveniencing a few billion electrons, after all. But whoever owns a piece of AI virtual real estate doesn't want the liability of transparency - there is the problem.
I am the heir of a few generations of East European shepherd and I can't even count the times when I've been arguing with the vegan community, because in some magazine articles they read about the harmful way of shaving sheeps. It doesn't matter that there are different climate reasons, breed and health related needs of sheep shaving. They have decided already, based on New Zealand research, that the wool production is bad and don't want to listen to the other side. I can't convince them not with my experience nor animal care degree. What if the whole information you get comes not from experience, but provided by AI generated answers, the market will pay to control the way that you are seeing the world, even more easily than commercials make it nowadays. Such a madness!
So glad you're covering this topic. As a medical professional I am mixed about AI, it seems to be both motivator and menace. I was trained to document everything and practice evidence based medicine. Now we have lots of ungrounded health information on the Internet. The fiber world has been my retreat. One thing your videos provide that i love is demonstration of techniques AND how they feel about in a tactile way. what you sense and the haptic experience. AI cannot do this as yet.
As a professional ghostwriter I’m skeptical and nervous about AI’s ability to paraphrase and not provide sources. How can we know if the material is accurate and if it should be cited?
AI can be a great tool for planning and idea generation but if you’re using it for writing it’s crucial to have it be a starting point or part of your research and not your finished blog post. As the community noted - the blog posts feel off especially when you know the author’s voice. Always add yourself back in. Thanks for talking about AI and addressing the pros and cons in both visual and written contexts 😊
In my opinion, AI is not scary. Instead, I think the people trying to control it and control others using it are scary. I also think that there will still be a market for people who value art made by humans over AI. I for example would prefer some types of art specifically because they are made by humans and if they were made by AI their value would be reduced.
A1 is too clever for its boots and I think once people have downloaded the app, they own't erase it so there - the creators have won and just like phones have become addictive, so too will this because humans are lazy and deceitful and will use it for the wrong things.
When I was a kid and watching The Jetsons, I imagined robots cleaning and doing heavy labor, not creating art and literature and eating bon-bons while we do all the cleaning and heavy labor. This is not the future I was promised and I don't like it one bit.
Every technology is only as good as the political and social system it is embedded in.
If this AI tech came into a world in which the main objective was to give every person the best opportunities for self-actualization, then it would be welcomed as a great new tool that could ease access to information.
But it comes into a world in which any artist or creator has to commodify their creations in order to earn the money they need to survive. And that's why it becomes dangerous.
plot twist : evie is a computer generated character
i think you hit the nail on the head with the points you approach, it doesn't come across as a moralising fearmongering on how computers ruin our lives but a pertinent detailed analysis at the effects AI is already having on our lives. A lot about it reminds me of the societal and mental health issues we've been experiencing since the democratisation of the internet and social media (and the industrial revolution at a large) I hope (though i doubt it will happen and if it does it'll be to ensure some people's priviledge) that we will have some regulations or transparency to help us really understand what we're getting into when engaging with AI
please keep on being an optimist, we are really in need of people like you nowadays
Is that a MingoAndAsho supported spindle?!?!?!?! I love mine. I still need more practice on it, but I absolutely love mine.
It is! 💜🧶
This was thoughtful and insightful. You were so eloquent with this conversation. Thank you
Post script. Ourtifact is an AI generated site to record a podcast which records your family history and experiences.🤔😳no, I do not think so. My albums, journals, and personal ephemera is the story. Passing down information, no matter if it is a memory or memorabilia , is a human relationship that has synaptic value. They are trying to retune our telomeres.
AI will have access to those stories. What will it do with those stories? I have so many concerns.
@@JillianEve Yes.Thank you Jillian for your presence. This is a deep concern of mine as humanity wanes. Stay close to the Earth and keep discerning truth.
Have a nice day all.
I like the comment it will not become accidentally benign. One of the things I think a good algorithm should Supply is references for the compilation it has generated. Also I agree that when something is repeated over and over it starts to become thought of as fact even though it is incorrect. Rather like the news. Certain things are repeated frequently until people start to believe even though they have done no fact checking.....You know with the medias non-voice writing style that is designed to enrage maybe it is being generated by a chat bot. 🤣
My granddaughter creates patterns for knitting and crochet.
This will affect her livelihood, won't it?
Unfortunately, yes. I think it will have a big impact on pattern designers.
I love watching your videos because of your talents with all and any imperfections with your presentations. AI is a program written by others. I don’t want to watch others.
As an experienced seamstress and fiber artist but primarily a person who works in the tech space that spinning like mad with AI. You are giving out a lot of misinformation on HOW to use AI. You actually used chatgpt like a search engine and your prompts were very misinformed to the AI. I promise when used CORRECLTY you can make great use of this tool. The tool is not the problem - the lack of understanding on how to use it is. If you unleash a person who has never touched wool and a give them a spinning wheel and almost no instruction they WILL use it wrong and make sad yarn. If you sit with them and TEACH them they will make lovely yarn in a short period of time. If you would like to contact me you are welcome to and i teach you about prompt engineering so you can learn and offer people a good idea of what the tech can do with proper knowledge of how to use it. I too did just like you in the beginning but then i went PAST my personal bias of the tool and did my own research. I am a product manager in my day job - my job is to make new features and progress the products i manage. We dont have a fine tuned model for spinning yet. I am actually in the works of making fine tuned models that would be usefull to knitters, spinners, and sewists. This is the early days, the dawn of a new era. Spinning survived the mechanization but i am sure the first years of mechanized spinning, knitting, and weaving machines felt like this to those spinners. We are at a crossroads and I firmly believe you must learn to adapt to not only survive but to thrive. I would be happy to have a zoom with you and record it to teach this community how to get great prompts that work for you, how to use ChatGPT for its best uses (it's not a search engine!).
Thank you for this perspective. This was my first week of really playing around with it and I am going to continue to learn more and see how I can use it in beneficial ways.
I have ABSOLUTELY NO IDEA WHAT YOU ARE TALKING ABOUT! What is AI technology???
Computer programs that are designed to simulate human intelligence.
'Liked' to boost the signal, so this gets seen... but in NO way do I LIKE the implications!
Thanks for starting this conversation!
As an aside, I ADORE the way you played with the transitions for each segment, to show us a glimpse into the Uncanny Valley. That was brilliant! I don't know what the Robot Plagiarists have in store for us - but I already hate it.
Thank you! Creativity is my personal rebellion.
@@JillianEve 🙌☀️🙌
Jillian , in conversation today with a friend, it was stated that chat gpt is very outdated. Interested, but not surprising.
“You will be assimilated. Resistance is futile.” is here.
Hi Evie, what do you think of adding a . . . I'm not sure of the word, maybe "certified", similar to food that is non GMO or meets organic standards; some kind of certification added to the blog post, or pattern, or book, or vlog/ podcast, that says "this information was created by a human and not with AI" --- something like that, I don't have it figured out and I'm not the one to figure it out, but within the Fiber Arts comunity, there would be a "certified not AI generated content" --- thank you for this awesome video Evie.😊
A very interesting and thoughtful video. Thanks for sharing this and giving us a lot to think about.
First of all, you have a wonderful sense of humor even when discussing a very serious issue!
I was using Dall-E a few months ago and it seemed like a total failure. I'm currently in Saskatchewan and wanted an image of a beaver in a Roughriders shirt mowing a law. It didn't even get the team colors right. Maybe it has learned what the first W in WWW means, perhaps not. I haven't tried again. I also tried a couple things in different artists' styles. Again, really bad. I eventually just wanted to compare a "Marilyn Monroe, Andy Warhol style" with the original. It looked a fair bit like her, but it also looked like she was kind of melting. I'm sure that it will get better because, as you said, it's learning.
Thank you for addressing this. I've never played with AI so I haven't given it much thought beyond it sounding scary.
I have a couple of friends that have recently had experiences with AI.
One is a Professer at a local University. She now has to use software to scan papers from her students for plagiarism and AI. Unfortunately, a couple of the AI papers would had slipped through if the AI's hadn't told on themselves in the last couple of paragraphes.
The other friend can't wait to get an AI program to argue with it. She loves to argue and most of her friends are fed up with her. I am worried for her. Worst yet, she has add and I am worried that she will be caught in a rabbit hole she will not be able to get out of.
Thank you for this video.
I have also come across students that have vented on the internet of never using AI in their work, and it got flagged for AI use, despite them recording their process for their writing to prove they were not using AI. Unfortunately it may become to the point where the line is indistinguishable.
The other downside of AI, is that it takes a LOT of water to cool down the servers each day. on tech sites such as Gizmodo, it's claimed that Chat GPT alone can use up to 700,000 liters of clear freshwater each day to cool it down, which can have its own severe environmental impact as well. One might think it's almost as bad as crypto mining.
@@kieraoona First, I'd like to correct some misinformation. GPT3 took a total of 700,000 litres of water to cool it's data stacks over the course of its training. This was an estimate for the total training period, not a daily use figure.
That said, this is not unique to training Large Language Models (aka AI). All data centres require massive amounts of electricity and water to run and maintain their stacks in the environmental sweet spot. Maintaining 68-71F and relative humidity at 40-60% is a water and power hungry business, especially when data centres are built in hot places (like Texas) or farmed out to less developed countries with less efficient systems.
"AI" is already ubiquitous, from predictive text and spelchek to cloud services and the mighty RUclips algorithm. Unless you totally unplug from the modern world, you are using "AI" on the daily.
Food for thought that's for sure. Thank you !
This was so well done Jillian. ❤
Off topic, but representative of language preserving the information of cultural relevance, Pandora’s Box was actually a **jar.** It was Pandora’s Jar. Large clay jars filled with food were buried in the earth… this is the original context of the story. If you like ancient history and mythology it’s a really fun rabbit hole.
Of course it would be a jar! 🤔💯🏺
Sadly, this is an almost inevitable progression in the use of information. Computers have made it possible to aggregate and derive statistics from large amounts of data for a long time. But even before the widespread use of computers, 20th century publications like industry trade magazines shared information on trends in purchasing and on how to capitalize on those trends--for example, how to create new food products to appeal to consumers and how to market those products in the most effective ways. AI is not really new. Its data-gathering and data analysis ability is just far more extensive than before. In essence, it's one more step along a path (maybe even still a sub-conscious path) to smooth out variations, remove uniqueness, reduce choices. I don't think there are philosophical or political goals here, only financial ones. The more that choices can be limited in any industry, the lower the cost of providing the products and the more profit there is for the manufacturers and retailers (and for their investors).
The growing use of AI in the arts is very similar to how manufacturing and retailing in general have changed since the beginning of the 20th century. The small locally owned businesses (often with unique regional products) have been almost completely replaced by the 'big box' stores with the same products in each one , Our food choices are far more limited than they used to be. Almost all food comes from a relatively few producers, whose products are heavily (and often deceptively) marketed. Many of us have had the experience of bringing some item of produce to the checkout line in the supermarket, and being asked in all sincerity what it was. Rhubarb, radicchio, shallots, you all know what I mean. Blank stare from the cashier--never heard of that before. That's not because the cashiers are stupid. It's because the variety of food choices that people have been exposed to has drastically decreased over the last 100 years. Everyone wants to see some kind of global conspiracy in this, but the reality is probably that the bottom line is simply profit. Whatever can be done to reduce the cost of products and to induce people to buy more of those things increases profit.
So what does this have to do with those of us who practice the fiber arts? As you pointed out, one thing it means is that we will have to be increasingly careful about who and what we believe. But to me, it also demonstrates the need to practice those arts in public, to promote them wherever possible, to teach others at every opportunity, to be openly and proudly unique. We never know who we're going to affect.
Thank you for bringing up this topic.
Honestly I feel like "we should be careful to give quality things to the thief stealing from us" isn't quite the solution here.
But also there needs to be a real change in how we look at "content creation"
"Content creators" are what we are to the big companies profiting from us, and I understand that using it as a self-descriptor was probably to try and put some pressure on them because "we are the ones who MAKE the data you make money from" is a valid argument, but... Well, they sure found a way around THAT, didn't they? When art, education, discussion etc are seen as "content," of COURSE it doesn't actually matter if it has truth, context, emotion, or if it's moral and lawful (and by that I mean "doesn't steal from others and doesn't encourage violence against the vulnerable," before anyone thinks I'm supporting said violence lol). And playing into that, calling our art, our work, our important discussions "content" is encouraging stuff like this and also leading so many artists to burn themselves out chasing algorithms. We need to stop calling it that and seeing it as that.
We are artists. We are educators. We are community members. Computers (in reality, assisted by thousands of underpaid workers, btw) can make "content," because content is just data for people to passively consume. And who cares if that data is just a HD collage of stolen things that may or may not be good or true?
People are being so blind to what this AI tech means. They (often) misunderstand where it comes from, where it gets its info, how inaccurate it can be (BARD anyone?). There is all sorts of info on the internet - which the AI's pull from - bad info can be given as easily as good info. Add in any programming bias (because no matter what, programmers are people and people have a bias). People need to take this seriously.
AI not referencing specific sources is a huge issue for credibility. What will we call new falsehoods generated by AI? ‘Old chip tales’? Quantum fables?😂
You are brilliant! And courageous! And extremely talented!! Thank you. I love your lectures!❤
This topic is scarey. So if someone needs to write an essay, say their degree/study. Would it be considered plagerism as it hasn't been written by them? Thinking back before I retired 2016 from General Nursing. We had asian students who could barely speak English. But the essays they wrote were so well written and researched??? They either had another person write their essay or perhaps AI??
Legal Eagle, a law youtube channel run by a lawyer, did a video about lawyers who used ChatGPT to create actual case paperwork that they then submitted to a real judge in an actual case. They ended up getting into a lot of trouble because the AI cited a bunch of fake cases to support the argument the lawyers were making and the actual human lawyers just assumed the AI was correct and accurate and so didn't check its work before submitting it. The judge, also an actual human being, however did check the work. He didn't recognize any of the cases and so he looked them up and found that they were fake. They didn't exist. Obviously, he was quite annoyed and that is when the lawyers who submitted the paperwork said they had used ChatGPT to generate it.
My thoughts on this: it should inspire people to go back to doing their own research the old fashioned way:: hit up libraries, museums, other live people doing the thing of interest. From this presentation, AI is only giving an "average" of things found. That's not sufficient for the reasons noted: there's a major loss of details, heritages in this case. The future will require Intelligent use of mortal brains with the side input of AI. As it's been noted: garbage in, garbage out. It will be up to the human element to discern relevant input and not accept only what AI offers. As for those who merely want to make money until a thing goes dry on them: those types will be everywhere, no matter what. For example, w
witness current corporate attitudes that are affecting the environment. Such types will also need a functional human brain to spot and avoid.
Thanks Evie for making me think. I haven't knowingly used AI as yet but l remember a quote
... don't know the source 😂 ... oh yes my mum. 'Believe half of what you see and nothing of what you hear ... or now read.
To be yourself think for yourself. Interesting discussion and my thanks for starting it on your channel.
I asked Chatgpt to write me a knitting pattern for some handspun that fell between standard yarn weights. I told it my size, the wpi of the yarn and various design elements I wanted - it wrote a pattern for me within about 2 seconds which astounded me. I then asked me to write a "David Gilmour guitar solo". Again, I had the guitar tab in a second or two. I played it back, not really expecting it to be too good or, indeed, anything like David's style. I was wrong! It could have been a "lost" Gilmour solo! It freaked me out so much that I haven't used it again. (I kept the solo though 😋)
A very thought provoking and timely post. Many of these thoughts and arguments were popularised in the late nineteenth century too, through the Arts and Crafts Movement of the likes of William Morris et al, and concerned with Industrial Revolution’s effect of rendering the works of crafts people unnecessary.
One thing that has always bugged me is how, in children's books, spinning wheels and the spinning process is usually portrayed wrong. The wool does NOT go around the big wheel! This proliferation of inaccurate information will skew results, and as you mentioned, reinforce its' own mistakes.
This is wild and scary for our future. Technology is wonderful- it can be the best tool ever. But money rules the world, and that’s a fact. How very overwhelming
As an artist, I am not thrilled with AI. Especially when it comes to copyright. As a graphic designer as my profession, I am not allowed to just any image. I have to use images that are either free use or come feom a company that provides them to me for a fee. AI needs to be held to the same rules. If I was to write an article or even just a paper in school, I had to cite my sources. AI needs to follow these rules as well. As a starting point for a topic, like a prompt, it is a good tool. When I have an idea of what I want to paint, I don't simply look up a similar painting and copy it...that's called forgery! I research and look at various images. Like in your example of a medival woman spinning in a field with sheep, I would resaerch and find reference images for clothing, hair styles, the type of spinning done in medival times and sheep. Only then would I feel ready to paint the scene. But the resulting artwork would have something more...elements of me in how it was painted, the color choices etc. AI takes that all away. Though, AI may make it so that actual human created things gain more value. But things will have to get worse before it gets better I fear.
I know AI exists but I didn't know how extensive it is or its implications. So thankful you are REAL and not a generated AI persona. ❤
Thank you for making this video, Jillian! Just like any other computerized functions, if we feed AI garbage, it will give us garbage in return. Garbage in, garbage out! I agree with you that we need to be diligent about putting accurate and complete information in any of our online content. If we use AI for fictitious endeavors, we need to specifically state that the scenario is fictitious. AI is like a toddler learning new things and trusting everyone. Tale care to teach it correctly, and it will grow to be a kind and helpful technology.
I have noticed a tremendous increase in blogs generated by AI. When I search for ‘how to’ do whatever I often find that several of the top hits are AI generated. At this point in time I can recognize them by little clues. The voice is often the most obvious. The quality of the information and the supporting rationale are always lacking. But this is now. Over time these flaws will be reduced and people will begin to have difficulty differentiating from human generated content. That will make us, and all information seekers, susceptible to the mother-lode of misinformation! The controllers of the AI will be able to influence and eventually control our thoughts and beliefs. I find it terrifying.
Very well written. There’s so many things that I hadn’t even considered. But of course if it has access to everything online there’s no control over the output. The lack of sources is scary. I’ve seen people in the sewing community make clothes from AI. I’ve heard of people getting it to generate patterns for knitting as well. It’s alarming that the information given you was just slightly out of focus in terms of accuracy. We already have problems with this thanks to the Victorian re-write of history. Without the results now being slightly off kilter. You’ve brought up a lot of things that I hadn’t really thought about. It had never accrued to me that it would be miss labelling thing and not checking or showing sources.
Shannon Makes sent me and I will have to thank her for that. At present I don’t spin or weave but I do knit and have friends that do spin, weave and felt so I will send this on
My thoughts, first copyright: if the model for the ai is trained off your own work then you own the copyright, of the model is trained from online data from many ppl, then it cannot be copyrighted bc sources cannot be cited; second, sorces: we need to not accept ChatGPT as a scholarly source, similar to Wikipedia, I believe its ok to use ChatGPT as a starting point, but it needs to be proof read, fact checked, and accurate sources found for what is presented.
I think this is very scary 😮. In the pasted i learned how to do things from a community of people who got together every other month as a group. It was on hand learning from older people doing this for years. When this is from the computer, i can not touch the yarn someone made with their hands. I feel 😪 it's robbing us to see things in personal. I mess badly 😢 😭 😫 😩 my spinning group i meeted up with every two months. I like what you had to say 😏.
As an artist, fiber content creator and sewist, I downright refuse to use AI wherever possible. It has its benefit for some, but I could see it having a lot of negative impacts on how we express ourselves, or as a tool so people don't have to pay people for work needed in specialized fields such as art and design, where artists are already severely underpaid for their work as it is.
As a person that used to work in the advertising industry as a content writer and social media administrator, I would have killed for something like ChatGPT back then. NOw, after listening to your take on it, I think I might be a little more worried about where we go as a species moving forward and keeping cultural traditions pure and alive.
I completely agree. As a neurodiverse person myself, I have many worries about this too. These racist ideas from an AI that was supposed to know everything can be very harmful to those who may not know the truth. Great job with the video, from another yarn spinner.
this is onhe of the BEST discussions on AI that I've read so far.
Oh my stars! I had never heard of these AI tools. Yikes. You are right about fact checking. I am not sure how much fact checking is done even on a google search usually.
This technology is super creepy to me. I’m not sure what I would ever use it for so it could have a benefit for me, so I have no plan or desire to use it. I guess I just don’t like how it is now impossible to verify what is true or what is computer-generated. Like the self-checkout at stores, this is just gonna lose people jobs as well.
As soon as i have about this sort of thing some years back (ai) I strongly believe to the opinion that just because someone can do something doesn't mean we should. Unfortunately people in general don't hold to that opinion and that is one reason we've ended up with such things as this. Computers have their place but people are better all ways around so keep to people... Sigh. Not going to happen
I love all of the points you brought up in this video. You've been eloquent about the issues with AI, and especially the issues with art/content theft and sponsored content creation appearing to be earnest and instructional. I hope everyone in the fiber community sees this!
Keep in mind, this is just consumer AI; what's being researched is a bit more powerful.
Thank you for heartfully reporting this information. You have echoed my thoughts.
thank you for doing the research and providing this information. Creators are in deep trouble, obviously, with AI, in terms of protecting their copyrights. Yet, consumers of information from the internet are going to losers as well as we will not be able to trust the content (was it simply a copyright violation or is it a total misrepresentation of the concepts).
This reminds me of the conflict between sall local retailers and the giant warehouse stores.
Edit: Also the dependability of Wiki.
I wonder how long before it thinks for you in every aspect of your life such as how you spend your finances, what you think, what you shop for, hiw you rsise your kids, where you go? My feelings are that it can be a detriment to us as it becomes more intelligent.
you are a wonder...I always find your videos captivating. you are like no other spinner, weaver, knitter, etc'er I found so far on youtube. I love the content and the honesty in witch you present that content. looking forward to the next video.
A conversation I listened to recently likened A.I. to a genie escaped from the bottle. It is an incredibly powerful tool, but we need to be very careful the wishes we make.
This video is excellent! It hadnt even crossed my mind that AI can mix and match cultures and misappropriate cultural traditions.
I think part of the issue with ChatGPT when it comes to sources is that it is meant to answer questions in a conversation format. If you try to dig in and ask it about where it's information comes from, it changes the answer depending on the context like someone would in a conversation. If I ask about technical information and then start asking about how the AI works then it will tell me that it was trained on sample data that it no longer has access to, but does not have access to the internet. If you ask about horses, or knitting and then ask for it to cite its sources it will point to general knowledge, and it might add the caveat that its information can be incorrect. If you ask about the data it was trained on, it will tell you that it was trained on visual and textual information, but if you ask about something visual and then get it to talk about the data it has access to it will tell you that it was not trained on images. Like a human in conversation, ChatGPT tries to give you the answer that you are looking for, instead of necessarily giving you a researched answer. If you ask it about copyright it will tell you that it does not know where its information comes from and therefore it cannot guarantee that anything it says is not an infringement of copyright if you use it. It does not seem to have access to its own sources.
I want to know why, when you asked where it got its info, why it didnt give you a list of sites.... a bibliography.... listed like ingredients. most info used to least. so then our slow brains could use it like wikipedia and let it be a starting point
This was a very evocative video, the contrast between the generated images and your own are perfectly chilling.
This is so concerning to me. You brought up things I hadn’t even considered. Thanks for this episode.
I completely agree with you. It’s concerning on multiple levels… on so many levels that we aren’t prepared to deal with but it WILL be prepared to deal with in a hundredth of the time that we can.
If we thought there was division and turmoil in our culture before, now we have the perfect storm.
I asked it for a knitting pattern for a hat that said “Made By AI” I have knit the hat and it is super cool, but to me, it can only do the mechanics of something, not the vision, passion, or intentions of the maker. AI is amazing but it definitely has its limitations.
Agreed about it lacking passion, but I'm sure it will improve on its ability to generate complex patterns. I have big concerns about that for pattern designers and tech editors. 😞
@@JillianEve I think tech editors will be safe for a while yet, *IF* those who hire them can be convinced that LLMs like ChatGPT really do need a human to double check its output.
I've played around with ChatGPT a few times, and it eventually fails even basic arithmetic problems because it's not actually calculating anything. It's just throwing words up on the screen that it's calculated have the most probability of being "right" for whatever it was asked.
Thankyou evie, i am so ignorant to technology, that was food for thought. Love your presentation .
Also, some of those AI generated images could use a body horror warning, eek!
Great conversation! I agree with your concerns and worry for the people who have put in so much time and effort to put together patterns only for AI to steal them 😢
This is a fabulous post. I had thought that AI had nothing to do with me. Now I see how wrong I was. Wow! I’ve got a lot of thinking to do.
😯 Can AI produce card weaving patterns???