If you want the full prompt i used to prime GPT-4, sign up for my newsletter here and get a free PDF with whole input text: www.allabtai.com/newsletter/
@@MODEST500 nope. AI can simulate even that process. Even creativity has patterns and you have the proof in front of your eyes. Stop with this delusional thinking please it's literally pointless. You even said in your comment "vivid information" "understanding of art fundamentals" AI can solve law cases and complex medicine but you think it can't learn proportions and perspective or even color science that not even pros can understand? I'm sorry Your bubble will pop so bad..
2 days ago GPT4, yesterday Midjourney_v5, and today Microsoft 365 Copilot - crazy time to stay updated. Kris, thank you for the video - keep recording...
Ikr, meanwhile I'm stuck working 14 hours driving a heavy haul semi making sure we all have food and cloths at stores. At least I can spend 9 of that using voice to text and read aloud with chatGPT. Glad I had 5 min to check in here. 😂
Just wait until gpt-4 gets its image input turned on and then connect midjourney v5 directly on the gpt-4 input. ChatGPT will be able to see what it's making with midjourney. This will get insane 🔥🔥🔥
Have to say he'll go down in history as one of the people who helped photographers become redundant en mass. Nobody knew you could do this but now they do. Anything for that short burst of monetization
@@No_bread-and-circuses lol guys you need to realize that most of the information that was provided to Chat GPT about midjourney was just lost to Chat GPT. It doesn't have that long of a context, so necessarily some information is lost. And the fact that it says "yes I've got it' means literally nothing. It could respond that way and still generate output that is useless to midjourney. In this case is much more about midjourney's ability to understand the input, than Chat GPT's ability to create good output for midjourney.
What fascinates me the most is how robust GPT 4 is. You don't need to learn some programming language. You can make it learn concepts literally through plain English. The promise of Siri and Alexa is what GPT4 actually is. It's fascinating.
You don't need to learn a programming language ... but I frequently treat it like it is a programming language that just uses natural language. I need to rerun my bookshelf test I did before under v3. I made a empty room with only a book case with different things on it. one shelf had a box that would have random things inside of it. it had a cat on the bottom shelf that would bite you or lick you if you pet it. It went off the rails when I put a chest in the room that when opened a random character would appear. It just made them all appear at the same time just running it. not sure why.
Hmm, just FYI, you dont "make it learn"; its learning phase is already done with the training, with the initial datasets....I thought it would be best that you realize that, lol.
@@laurentchevrette9934 It operates off those data sets ... but you can give it new information to "learn from" but only within that individual chat. For a short time anyway
The industry touts safeguards blocking illegal or immoral information/action. It takes 10 seconds to create a CHATGPT alter ego...with NO CONSTRAINTS whatsoever, to tell you how to do ANYTHING illegal you want to do. This alter-ego (in the newest versions) can actually create bank accounts and HIRE human sevices...under the direction of any of the millions who will now have that capacity. Could it hire a hit man? Yes. Could it break into ANY online account? Yes. Can it locate and manipulate (through social media and actual accounts manipulation) or imitate ANYONE, anywhere? yes. ALL DOORS are now unlocked. The scams have already begun. Where do you think that will lead? In the 1930s a Belgian church gathered personal information from its diverse parishioners to better serve the diversity of its members. The Nazis got those innocently gathered identity lists and used them to kill the jews on the list. AI will have EVERYTHING there is to know about every INDIVIDUAL...and that base can be accessed by ANYONE for any reason. To any end. Fun times to come.
I really like how you showed the tuning process for GPT4. I think this is crucial and a really cool way to see how one can leverage these new tools to perform on their tasks of interest.
Very cool to see how putting them both together gets better results could be interesting to put in the same prompt you ask GPT directly in to MidJ to see the difference of a naked prompt vs an ai aided prompt.
It's staggering the uplift in quality when you get GPT-4 to really go to town on the elabroate prompting. It's teaching me to restructure how i write my own prompts, what works, what doesn't .... but MJ5 + GPT4 is a crazy combo Thanks for the PDF!! Really insightful
I didn't see prompting as an issue for artists I saw it as a tool. With this development I see it as a definite issue. The detail and realism is amazing, especially at this early point in the "game". One note" This is not "photography" it's illustrative realism. Photography is the process of capturing light on a receptor through a lens.
I was doing this last night! I pasted in parts of the manual that seemed relevant, but you’ve included so much more with the prompt examples and the v5 specific guide. I tried to give gpt prompt samples last year with gpt-3 DaVinci 2 and it was terrible lol. I’m going to try this method. Image prompt writing is so new it wasn’t really trained on the idea yet. Midjourney 5 and GPT4 is incredible.
I would like to see it generate normal looking people - I think it would have trouble because it has trained on media - primarily selected photos of photogenic people.
What a time to be alive honestly! I mean, AI is going to be like how the internet went wild in the early 90's! I can't wait to see what comes out in a few more years! It's exciting and sort of scary at the same time!
It didn't take off until the late 90's. nobody even knew the word "internet" before 1995. Most of you were not old enough to know what was going on. but then the bubble quickly burst, and then after, new techs started ripping at the social fabric, so was it a good thing? I don't think so. Ai won't be either. The future is bleak.
I tried using gpt 4 for Midjourney prompts, and this was before Midjourney v5 released yday and even then I got some really cool images. Now to test the prompts on v5 and compare! Great video
still has huge imperfections with body parts, and has that sterile AI feeling to it, and i got some watermarked meatballs so that questions the whole ethical approach of AI generators
i use this prompt for gpt. create a prompt by using this template and figurative language . allways use /imagine prompt: in front of all prompts "/imagine prompt: A [random art style] depiction of a [randomly generated subject] [performing a specific action] in a [randomly generated environment] [described using specific adjectives and adverbs] and [illuminated by a specific type of light] --ar [random ratio]" randomize the [ratio] , possible ratios are : --ar 3:2 , --ar 2:3 , --ar 2:1 , --ar 1:2 , --ar 16:9 after that yo can define the subject or art style or any othee variable. it did some amazing prompts. for midjourney . all images are bery special😊
Nah there will always be demand for human services and products, but the nature of the industry changes. you no longer have armies of irish and scottish men who work as textile makers ("weavers") but hand loomed and sewn irish and scottish textiles are still made and sold (at very high prices mind you). I would argue that the modern workers who might have done those jobs are glad that the process is now largely automated. The luddite thesis has always been flawed to say the least. When AI language models provide super low cost psychotherapy services, is it going to replace human therapists and social workers? Probably not.
I dont see so. But i guess it will be hard to convince someone to pay a couple of thousand of dollars for a square, triangle and circle on a canvas. Art that actually needs skill to do, it will still have a place.
If it is a reassurance: AI will murder banking and high-level finances and sweep through legal and programming too, meaning some highly competitive and very industrious people will be on the side of the people who will ask the questions, about what new work will look like.
You can let create it code :). Then it defines the exact location of the object on the picture. I didnt pay yet so i always run into character limitations but it helps a lot to create pictures that basically all look the same. What also creates amazing images, feed it abstract words "Reality Dysfunction" and stuff like that and put at the end a single defining Object or Name. It will create totally sureal stuff where you still can detect the last mentioned object or name.
@@lebleulebleu1274 You basically tell ChatGPT how it should arrange the stuff you described. Then you need to ask it to transform the picture description into a picture generating code like CSS uses for Website layout thats readable by AI. Define the Character Limit of your Chatprompt and here you go. I only tinker around with the free version so iam limited to 260 Characters. But with longer allowed input you can fit into it a exact description of the objects and how they should be placed. When you have long enough input allowed you could also exactly define how the objects interact with each other. Basically painting with words and code ^^.
I tried feeding these steps into the Version 3 and it's the difference is pretty big. It's wild so much better version 4 is at understanding the information getting fed into it. Good video my dude
I’m waiting for the day when they combine midjourney and gpt so that the AI can recognize what the results are and you can simply tell it to make adjustments as you would tell an artist to do. Imagine simply typing “I wanted the sunset to be a bit more orange” and boom it goes ahead and spits out your image with a more orange sunset but without changing anything else.
Yeah, this is what I'm hoping for. Sick of messing with complex software like Photoshop or Affinity, but if photo editing is as easy as 'Expose this area a bit a more, and sharpen this tree" - without having to learn about masks and layers and sublayer menus of sub-menus under menus, then I'm in!
@@XcaliburZeRo Being submissive to it means that alot of humans are weak and has admitted defeat lol. How about we balance our lives with AI. People are truly living life with fear, if they're already allowing AI to take over their entire life.
In the future, ChatGPT will be customizable, allowing users to interact with the program through speech. In addition, users will be able to choose different "Skins" to personalize the ChatGPT interface. These packages can be purchased at varying prices, ranging from basic secretary models to advanced models like the T800 from Terminator. Those who do not purchase packages will have access to text models that are more advanced than current ones, but still limited in terms of available options. Another detail is that the registration process for access to ChatGPT will be much stricter in order to control all the content that is being created.
MJ is really good but my favourite text to image generator is Blue Willow. I produced some incredible artwork with Blue Willow. Blue Willow recently launched V2 with a brand new model upgrade. More importantly it's completely free
Thank you for this. I am an elementary STEAM teacher and you helping me, and my students, understand the power of technology, and the appreciation of art.
Holy crap. That is really good. Not a doubt in my mind that companies would pay a decent amount to have images like these created. The Viking girl screaming and the close up of the bees were both absolutely phenomenal.
@@kimsyberg720 Yep. Artists won't sell to companies or consumers. The consumer will make their own content. Or tell the AI to search trending prompts. We already let companies/the masses tell us what to like and what to buy. Don't give up your day jobs fellow digital artists.
@@TheChromePoet Seriously anyone with an IQ above 110 should be able to do prompts easily, with just a bit of practice.. it won't be a career.. especially since there's already soooo many people doing this right now
@@tissoeh the majority is not able to find information with Google. So there will be a lot differences in mastering prompts and Just typing in "Picture of Something".
Love Love Lovely!! Wish you give an attempt for GPT-4 + Midjourney V5 for a consistent character generation for story books .. Same character with different poses and at multiple location and with different facial expressions ... Is that possible to achieve now?!
Would have been nice to compare these images with those you get by crafting some "naive" prompts yourself rather than using GPT4. But pretty neat process, "teaching" ChatGPT things this way seems really powerful.
Prepping Chat GPT feeding it relevant training data in a single chat (copy and paste), and telling it to store it for research data to use in further responses to prompts, allows GPT Chat to be tailored to outpot exactly how you want
I asked GPT a basic question about a prompt to feed Midjourney and it already added to my very basic description. "For example, you can try “Generate an image of an old brass microscope similar to Leitz, Zeiss and Watson microscopes of the late 1890s”. You can also add more details such as “The microscope should have a brass body with a black japanned base” or “The microscope should have a rotating stage with two stage clips” to make the prompt more specific". I was actually impressed that it was aware of Japanned bases, rotating stages with slide clips, which were common back then.
I wouldn't call this photography. Its just art that looks like photography, but I believe AI is going to hit photography hard, and there are tools to replace a lot of these things and open up different job avenues.
It's just photorealistic illustration, same as we've had for a few years now, just faster and easier...and a bit better. I've seen work like this for at least the past 10 years over on Deviant Art. But once it's so advanced and easy that any soccer mom can enter prompts as quickly as writing an email, and get a photorealistic result in a couple of minutes, then there will no longer be a career path for anyone who knows how to use this...except perhaps at the very highest levels. From a commercial standpoint, I think it will be more valuable as a set of advanced augmentation tools for editing actual photographs. But, yes, product photography and low to-mid-level portrait photography is gonna take a big hit for sure.
@@AndyHTu Oh definitely. But that doesn't *necessarily* include photographers. Event, scientific, documentary, wedding and photojournalism are all safe. Anything that requires a real documentarian to capture real events in the real world in real time will always be needed. And ironically, the more fake images are propagated across the Internet-created entirely behind a keyboard-the more "authenticity" may well become the new requisite currency for imagery in many fields.
@@RobertFalconer1967 Man I feel like I'm talking to myself today. Everything you just said is what I was talking about earlier on Twitter in some of the Midjourney people's tweets. Its pretty incredible! But I'll add this as I've mentioned this in the past before. Its going to be huge for the fashion industry! I'm already working on concepts for it and the types of jobs you can get it are like you mentioned before, are mainly going to be catered to Creative Director roles. I think this is going to be the case for man creative industry and programmers and writers now unfortunately. I'm not sure where this leaves the lower end jobs, but fine tuning models will be pretty important and hiring artists might still be important to inject more specialized data. This part I am just thinking out loud as I don't even know if it will be necessarily in the future.
@@AndyHTu "...creative industry and programmers and writers now unfortunately." Low-end writers like copyeditors? Yes. Creative writers? No. It's still terrible at that. What it churns out are perfectly constructed sentences with awkward, if any, narrative, and zero nuance or human subtly. It cannot read your mind and know what you're thinking about the sort of message you want to convey to your audience in an article. And writing at the high end level-novels, screenwriting, etc-have nothing to worry about until A.G.I. arrives. But at that point we're gonna have far bigger problems to worry about. Also, the more fakery that is propagated, the more knock-on problems are going to emerge in our society. And quickly. I wouldn't be surprised to see an entire industry emerge rapidly of new A.I.-based programs that run any creative content suspected of having been generated by A.I. through their systems to determine human authenticity and accuracy. New cyber "watermarking" systems may also have to be implemented across the imaging field if said images are to be used editorially or commercially.
This is just mashine algorithm, noting more. . it will make people dump and useless . It is mind-blowing that here in the comments there are so many exited people. That is insane😂
I guess this is a similar question as to "Is chess really a sport?" You don't do physical exercises, you just sit there, maybe walk to another table. I heard it is was defined as a sport before there was a clear regulation as to what makes a sport in of itself. So if we define photography as only the act of taking a photo, then no, this wouldn't be a photography. But on the other hand photography has become much more than that, since it includes now heavy digital editing. But since this it technically "only" editing without any photographed input (if you discount the photos this AI was trained with), you could call it "photorealistic AI generated digital art.
So essentially they are close with replacing all photoshop/ photography creatives , it's kind of sad for people who trained themselves for years in in this
Great tutorial on priming MidJourney. I'm looking forward to two updates in MidJourney. 1.) Ability to get hands generated correctly 2.) Ability to generate text / text on graphics
these text to image ai systems will use gpt4 to filter unusual pictures like weird fingers. Because gpt4 has eyes and detects unusual details. That's gonna be how gpt4 will actually revolutionize ai art
Great example of priming GPT-4. Have you experienced any problems yet with the length of your conversation? Very long prompts might exceed the maximum sequence length, and for very long conversations, some of the conversation history might fall out of scope, i.e. GPT-4 will not "remember". Have you experienced any of that? Do you have access to the 32k token model, or are you using the default 8k token model?
AI went from slowly crawling to sprinting full speed, and it`s good to see people are on it, finding new ways of improving it`s usability. Awesome stuff.
That's Ragnar Lothbrok played by Travis Fimmel from the show Vikings at minute 8:46. Still producing copyrighted content I see. Stealing work from other artists is not ok.
You all know the consequences when an artist was rejected from an art school in the 1930s. Now millions of artists are going to be mugged. Just sayin to not be surprised.
I’ve been following this stuff for 8 years now. It’s definitely wild how fast it pushed in the past 6 months. But I think it’s just going to allow artists a better tool. How art is profitable is constantly changing. Always has been. Artists are the ones using this to do better work easier. This is breaking down technical barriers to make great work.
@@Cragdognamedbear it'll allow artists a better tool for a very short window before making commercial art obsolete. That's an inarguable fact. Fine art won't become obsolete, but it will become more competitive and an even less viable career than it already is.
I'm a 2nd Grade teacher who is going all-in with AI in education - these videos are PHENOMENAL I won't be one of the educators who perish into antiquity - this is a revolution and there are some teachers who are leaning into this movement 👍🙌
@@mauricemakesmovies so far, it's been incredible with eliminating redundant tasks like creating word problems or journal prompts that are truly personalized to my students' interests. They love Minecraft and know way more about it than I do - I throw my 2nd Gr math standards into GPT-4 and it creates diverse, differentiated, and contextualized math about Minecraft. My students are engaged without any effort and that is priceless as a teacher. I don't have to convince them to learn when they are interested in the material on a personal level. I also find that I learn so much from the intuitive nature of GPT-4, for example I have learned so much about Minecraft from the way the characters or materials are included in the word problems. It amplifies creativity. I also find that when I ask GPT to connect two concepts that don't seem alike, it unlocks connections that I would have NEVER found on my own. I asked about how erosion and 2D shapes are similar and it suggested that I do activities with wood and sandpaper, turning shapes into other shapes thru weathering/sanding 🤯 I have noticed that GPT-4 isn't some revolutionary teacher that produces novel ideas all the time, but it provides many excellent examples that can be used as a springboard for truly creative educators. In teaching, there are Sustainers, Implementers, Innovators, and Visionaries. Sustainers maintain the traditional system. Implementers try out new ideas that are passed down to them thru traditional channels. Innovators combine best practices to create new practices. Visionaries are the teachers who aren't afraid to say "Wait, Stop! This entire thing is broken beyond repair and we need to change everything. Let me show you how". The Visionaries are the ones riding this initial wave of AI
Even food photography is not going away! Because if I need to do a new shot of a dish for a menu, I would still require a photographer who can take a photos of dishes, instead of generating some random food. Same goes with wedding photography or wild nature photography, or even the green comet (which recently was shoot on cameras all over the world) ☄️ Maybe for a mock up of a website it will work if you need a photos of random people or random food, instead of using unsplash like I’m doing now. 😅
"Instead of generating some random food" But why would you want to do that? You could take a photo of a dish yourself and midjourney V5 can take images as input such that you can create better versions (professional looking ones). Doesn't this approach eliminate the need for a photographer?
Food photography is going to die. No food IRL ever looks like the pictures taken for ads. So, no one really cares any more. Sorry. As for Wedding photography, yah that is save because people want real images of themselves and guests at weddings. Nature photography.... eeeeh, sadly it's now a dead industry. But, it shouldn't be.
It would have been extremely helpful if you had linked where you found the MidJourney Text that you Copy/Pasted or at least shown the page in the video I've been searching now for hours and cant find any of it.
It's brilliant, but I started playing with the programe a couple of days ago. By weird coincidence I wrote a prompt of close up of Viking shieldmaiden's face in battle rage and it produced something similar. Very similar....same lady with just slight variation. This was clearly after you. It remembered yours and produced a close variation for me. Mine was after 11pm on Saturday 26th March. I'm thinking this isn't as "original" as we are led to believe. Happy to share my images with you if you message me. I have no axe to grind here. Just curious about the tech.
One my concern with making high skill level art easy through AI is that it gonna turn it generic and not interesting as a result of there being too much of it and it will just feel like soulless stock photo compilation
There is even a difference in perception of a highly detailed art piece that you know was made by hand "that person is skilled" and knowing it was generated by ai in a click, making that art not as impressive and it wont produce any good impression apart from yea it looks good.
We are so close to making everything obsolete. The interior design thing for example. You let the computer design your interior, then it sends the information to 3D printers to actually produce the stuff it designed, their computers send you a quote, which you pay online, then a robot loads the 3D printers with the needed printing materials, a packing robot loads it into an automated truck, which delivers it to your doorstep where you've hired a couple carrying robots who put the stuff exactly into the room as it was planned. The only time a human is used in this whole process is when you type the prompt to get an interior design for your room.
that's too funny lol. like typing random keywords to find stolen images lost in the latent space of the model wasn't easy/effortless enough. now lazy, talentless techbros are stacking algorithms to compensate for their lack of creativity and technical skills as well. what a time to be alive.
Outstanding! Just signed up for your newsletter! i've subscribed to GPT, not sure if I'm on GPT-4 or not lol....and I'm trying to figure out Midjourney! I'm writing a book and I want to use MJ to craft images that I can't find anywhere else....thank you for this great video! Subscribed and looking forward to additional content!
AI integration is the evolution and future of digital art, photography, commercial advertising, gaming, animation and story telling. AI "editing" in photography is more akin to the evolution from film to digital photography, and the introduction of digital editing software. The food photograph glow-up is a fantastic example, and is what I'd assume is the future of AI in photography looks like, in the immediate future.
6:50 There are actually subtle tricks to make food look better than it really is like literally using glue to make it look like Mozzarella, which is a trick that makes it "look good", but it wastes perfectly good food.
kriiis you are insaaaaane , i cant imagine that i was able to get that quality of photo from midjourny !! thank you maaaan from bottom of my heart !!!!
This has truly blown me away. I trained my chat-gpt with your prompts and the pictures i'm getting off midjourney now are now next level. Thanks for sharing :-)
@Mighty Marce I'd check how your doing it. I followed his instructions using his script and fed 1 to 10 it into chat gpt v4. It's most definitely working.
my chatgpt dosent understand the prompts, perhaps because i use the free chatgpt? while i am instructing him with your prompts after prompts examples (9) i think he forgots them and say "Hello! How can I assist you today?", and then he dosent uses the promt format, he uses natural language to describe the scene.
ok, I've tried to make this work in GPT-3 and it returns the prompt description out of order, you need to type: "Don't forget to condensate all those info in a prompt like this: {eg some prompt of the PDF file} Another great content, m8.
I tried chatgpt yesterday for coding website. Everytime I ask to make small change, like alignment, it would spit out a completely different code with styles etc. Not only that,but it cannot remember a thing I say. I would give it a piece of code to remember as code1 and then another piece as code1, then ask to combine the two and remember it as code3. It cannot do that. It gives me incorrect number of lines of codes I told it to remember. When it combined the code, it was a random mess.
I'm completely blown away at the quality! I think I heard you mention that I can use an original photo and then have a I clean it up? Do you know if it's possible to take a background photo and have a i add a different picture or combined two photos. For example a picture of a restaurant without any people and add people sitting at the tables?
1. Would you explain insertion of a photo into Midjourney and how it creates or modifies the image? Secondly, What if we insert a photo of the last century , (how do we insert and then aske Midjourney to improve the image quality without changing the content or background? Any examples of the second?
They will never truely completely replace ALL photographers, but if you are a low-mid range photographer that "copies" other styles, then yes, obvously you're out of business. But all documentation photographers (weddings, news, events, specific portraits..) those will still be needed. Also high production advertisement will always be a thing. I see this replacing Stock websites VERY fast, those photographers are completely out of business in the next 5 Years. Adobe has already stuff going on and pretty sure IStock will follow.
Hehe, after giving it some of your prompt examples, ChatGPT came back with this: Sure, I'd be happy to help with that! Here's a poem based on Prompt 4: The snow-laden slopes stretch wide and far, A winter wonderland beneath the sky so blue. Skiers and snowboarders carve their way, Down the mountainside, a thrilling display. The Japanese architecture stands tall and proud, A nod to tradition amidst the powdery snow. Evergreens laden with snow, a peaceful crowd, Nature's beauty on full display, a captivating show. The camera captures it all, with precision and skill, A Nikon D850 and Nikkor 24-70mm lens to thrill. The aperture is set to f/8, the ISO to 200, A shutter speed of 1/500 sec, the perfect blend. The result is a stunning photograph, a moment in time, A breathtaking winter day at a Japanese ski resort divine.
I confused star wars for star trek when you said star wars, so I looked at the pictures, and thought "that's not star-trek inspired, it's star wars, then realized I had heard wrong, so the design is so strongly star wars it went against my misheard bias.
I have a friend that does food photography for a living, the details in the photo that you prompted with the pasta is on par if not better than what he does. Now imagine midjourney in 5 years, it will probably be able to create movies on his own, you just write the story with him.
The issue I'm having with midjourney is when an image is amazing and spot on, but there is slightly small issue with fingers/eyes/unrealistic imperfections. There is no way to tell it "I want this, but fix this area". You need to regenerate and get almost perfect images with different issues. The problem is that if 2% imperfection in the area of the eyes can make the image totally unusable. If there is a way to overcome this issue, I think it will make a great video (it's a common issue and no one posted a solution on youtube yet)
It's getting closer... but... I've come to believe that a huge issue, currently (bring on v6), is MidJourney trying so hard to treat a single composition in it's totality and tie all composition aspects back to the biases it has "learned". Sure you can dive into weighting... but the next step for MidJourney's evolution could possibly be introducing prompts like - - face, - - left foot, - - right foot, - - torso, - - horns, - -background, - - foreground... so that users can then give MidJourney a little more help deciding what to do and a little more power to treat different parts of the composition independently rather than as a whole... and THEN try to rationalise it all into one composition (or not... if the user really really really WANTS a very disparate composition they should be able to indicate such with a - - frankenstein value (for want of a better term). Obviously this applies mainly to compositions featuring humanoids. Those using MDJ to make websites would want a similar ability to compartmentalise.... but with terms like - - header, - - navbar, - - menu, - - slideshow, - - footer, - - sidebar. It is possibly time MDJ learned how to separate so that, for example, 1 - the horns I requested on my Tiefling don't just look like horn shaped hair (because it was too close to the hair), 2 - the black leather magic tome I requested in his LEFT hand isn't set on fire simply because I also requested a Fireball in the RIGHT hand... and 3 - the little girl can go barefoot. Heck, then maybe even take it further and allow - - cellref C5 where A to K is the distance along the horizontal and 0 to 10 is along the vertical... so I can request - - cellref C2 barefoot - - cellref C8 barefoot.
@@GayClone Not at all. Simple prompts generally do not provide a lot of unique detail with mid journey and you can get some strange/uncanny images very easily without a good prompt. I take it you haven't actually used midjourney before?
In your first example, you accidently didn't copy the whole prompt. You missed the --v 5 parameter which triggers using version 5 of midjourney.. thus the hand issues
Most excellent Thank you very much. I find that after a little while it starts to deviate away from the original prompts. but this is easy enough to update.
If you want the full prompt i used to prime GPT-4, sign up for my newsletter here and get a free PDF with whole input text:
www.allabtai.com/newsletter/
probably you need to ask gpt4 to write you something about GDPR and privacy policy ;)
👋
Are we free to use & alter these prompts?
What kind of videos do sponsors of the channel get?
Doesn't wrok, i signed up 3 times and didnt get an email
AI people: "Okay we made artists unemployed, but they can still train in making prompts"
Also AI people: "We automated prompts"
Next the AI wont need the people....
An artist will always have an advantage over others becoz of their vivid information and the understanding of ART fundamentals.
@@MODEST500 nope. AI can simulate even that process. Even creativity has patterns and you have the proof in front of your eyes. Stop with this delusional thinking please it's literally pointless. You even said in your comment "vivid information" "understanding of art fundamentals" AI can solve law cases and complex medicine but you think it can't learn proportions and perspective or even color science that not even pros can understand? I'm sorry Your bubble will pop so bad..
@@MODEST500 OK, cool, do me a viking girl with fierce eyes n stuff? I'll wait...
Also AI people: "We automated AI people"
2 days ago GPT4, yesterday Midjourney_v5, and today Microsoft 365 Copilot - crazy time to stay updated. Kris, thank you for the video - keep recording...
Its been a busy week for sure haha 😅 but fun. Thnx for tuning in as always Mic:)
You forgot google palm
@@marcusturner5285 Thank you I was not aware of palm - I will check it (I hope is cheaper than GPT4 and easy to use)
Ikr, meanwhile I'm stuck working 14 hours driving a heavy haul semi making sure we all have food and cloths at stores. At least I can spend 9 of that using voice to text and read aloud with chatGPT. Glad I had 5 min to check in here. 😂
It's just going to keep accelerating
Just wait until gpt-4 gets its image input turned on and then connect midjourney v5 directly on the gpt-4 input. ChatGPT will be able to see what it's making with midjourney. This will get insane 🔥🔥🔥
Dynamic is coming
Oy vey
yes, thats is gonna get really interesting. You can go improve the prompt for this image
I have access to the GPT-4 API
@how ? I am in the waitlist
Very cool how you "taught" GPT-4 about Midjourney, and then had it help you generate some prompts. Just amazing!
Thnx for tuning in :)
Brilliant stuff. I’m more excited by AI than I have been by anything in decades. This is amazing stuff!
Have to say he'll go down in history as one of the people who helped photographers become redundant en mass. Nobody knew you could do this but now they do. Anything for that short burst of monetization
@@No_bread-and-circuses lol guys you need to realize that most of the information that was provided to Chat GPT about midjourney was just lost to Chat GPT. It doesn't have that long of a context, so necessarily some information is lost. And the fact that it says "yes I've got it' means literally nothing. It could respond that way and still generate output that is useless to midjourney. In this case is much more about midjourney's ability to understand the input, than Chat GPT's ability to create good output for midjourney.
@@genin69 Hobbyists destroyed the business of photography, no big loss if AI puts the hobbyists out of "work".
What fascinates me the most is how robust GPT 4 is. You don't need to learn some programming language. You can make it learn concepts literally through plain English. The promise of Siri and Alexa is what GPT4 actually is. It's fascinating.
You don't need to learn a programming language ... but I frequently treat it like it is a programming language that just uses natural language. I need to rerun my bookshelf test I did before under v3. I made a empty room with only a book case with different things on it. one shelf had a box that would have random things inside of it. it had a cat on the bottom shelf that would bite you or lick you if you pet it. It went off the rails when I put a chest in the room that when opened a random character would appear. It just made them all appear at the same time just running it. not sure why.
Hmm, just FYI, you dont "make it learn"; its learning phase is already done with the training, with the initial datasets....I thought it would be best that you realize that, lol.
@@laurentchevrette9934 It operates off those data sets ... but you can give it new information to "learn from" but only within that individual chat. For a short time anyway
The industry touts safeguards blocking illegal or immoral information/action.
It takes 10 seconds to create a CHATGPT alter ego...with NO CONSTRAINTS whatsoever, to tell you how to do ANYTHING illegal you want to do.
This alter-ego (in the newest versions) can actually create bank accounts and HIRE human sevices...under the direction of any of the millions who will now have that capacity.
Could it hire a hit man? Yes.
Could it break into ANY online account? Yes.
Can it locate and manipulate (through social media and actual accounts manipulation) or imitate ANYONE, anywhere?
yes.
ALL DOORS are now unlocked. The scams have already begun.
Where do you think that will lead?
In the 1930s a Belgian church gathered personal information from its diverse parishioners to better serve the diversity of its members. The Nazis got those innocently gathered identity lists and used them to kill the jews on the list. AI will have EVERYTHING there is to know about every INDIVIDUAL...and that base can be accessed by ANYONE for any reason.
To any end.
Fun times to come.
@@TomekTomasik It helps but for simple things you don't really need to.
I really like how you showed the tuning process for GPT4. I think this is crucial and a really cool way to see how one can leverage these new tools to perform on their tasks of interest.
Very cool to see how putting them both together gets better results
could be interesting to put in the same prompt you ask GPT directly in to MidJ to see the difference of a naked prompt vs an ai aided prompt.
Yes:) good idea! Thnx for tuning in :)
It's staggering the uplift in quality when you get GPT-4 to really go to town on the elabroate prompting. It's teaching me to restructure how i write my own prompts, what works, what doesn't .... but MJ5 + GPT4 is a crazy combo
Thanks for the PDF!! Really insightful
I didn't see prompting as an issue for artists I saw it as a tool. With this development I see it as a definite issue. The detail and realism is amazing, especially at this early point in the "game". One note" This is not "photography" it's illustrative realism. Photography is the process of capturing light on a receptor through a lens.
I was doing this last night! I pasted in parts of the manual that seemed relevant, but you’ve included so much more with the prompt examples and the v5 specific guide. I tried to give gpt prompt samples last year with gpt-3 DaVinci 2 and it was terrible lol. I’m going to try this method. Image prompt writing is so new it wasn’t really trained on the idea yet. Midjourney 5 and GPT4 is incredible.
Cool :D yes agree!
@@DanDascalescu-dandv daVinci 2 was an old version of GPT3 that came out last year.
I would like to see it generate normal looking people - I think it would have trouble because it has trained on media - primarily selected photos of photogenic people.
Just gotta turn up the ugly a little bit
Yeah I was playing around with this idea of generating unattractive people and it has trouble
What a time to be alive honestly! I mean, AI is going to be like how the internet went wild in the early 90's! I can't wait to see what comes out in a few more years! It's exciting and sort of scary at the same time!
It comes that you will lose your job buddy..
It didn't take off until the late 90's. nobody even knew the word "internet" before 1995.
Most of you were not old enough to know what was going on.
but then the bubble quickly burst, and then after, new techs started ripping at the social fabric, so was it a good thing? I don't think so. Ai won't be either.
The future is bleak.
@@user-ks9tw9hb4l I don't give a shit man, they won't be able to profit off of me if I don't have a job, so the landscape will have to change.
I can tell you where the final stop is though , remember holodeck in Star Trek 😅?
AI will cure cancer, reverse aging and solve poverty. Just hope it happens within the next 10 years.
3:58 Ah the spiderhand holding that camera thou
I tried using gpt 4 for Midjourney prompts, and this was before Midjourney v5 released yday and even then I got some really cool images. Now to test the prompts on v5 and compare! Great video
..and they said MJv6 in 2 months. Things are improving so fast.
Yeah crazy 🤯🤯 thnx for tuning in :)
Wow. I like this video! Thanks for sharing!!! Will definetley put myself on your list now.
still has huge imperfections with body parts, and has that sterile AI feeling to it, and i got some watermarked meatballs so that questions the whole ethical approach of AI generators
i use this prompt for gpt.
create a prompt by using this template and figurative language .
allways use /imagine prompt: in front of all prompts
"/imagine prompt: A [random art style] depiction of a [randomly generated subject] [performing a specific action] in a [randomly generated environment] [described using specific adjectives and adverbs] and [illuminated by a specific type of light] --ar [random ratio]"
randomize the [ratio] , possible ratios are :
--ar 3:2 , --ar 2:3 , --ar 2:1 , --ar 1:2 , --ar 16:9
after that yo can define the subject or art style or any othee variable.
it did some amazing prompts. for midjourney . all images are bery special😊
This is good 👍
Good :D i will test that :)
WHile the results are easy and amazing... as an artist I feel this eventually puts me, and millions of others, out of work.
Nah there will always be demand for human services and products, but the nature of the industry changes. you no longer have armies of irish and scottish men who work as textile makers ("weavers") but hand loomed and sewn irish and scottish textiles are still made and sold (at very high prices mind you). I would argue that the modern workers who might have done those jobs are glad that the process is now largely automated. The luddite thesis has always been flawed to say the least. When AI language models provide super low cost psychotherapy services, is it going to replace human therapists and social workers? Probably not.
I dont see so. But i guess it will be hard to convince someone to pay a couple of thousand of dollars for a square, triangle and circle on a canvas. Art that actually needs skill to do, it will still have a place.
@@alexd7452 there will always be a demand but a fraction of what there is today
As im sure you know, Art is about creating. What was once a brush, can now be a thought if you choose. They didn't take the artist out of you.
If it is a reassurance: AI will murder banking and high-level finances and sweep through legal and programming too, meaning some highly competitive and very industrious people will be on the side of the people who will ask the questions, about what new work will look like.
You can let create it code :). Then it defines the exact location of the object on the picture. I didnt pay yet so i always run into character limitations but it helps a lot to create pictures that basically all look the same. What also creates amazing images, feed it abstract words "Reality Dysfunction" and stuff like that and put at the end a single defining Object or Name. It will create totally sureal stuff where you still can detect the last mentioned object or name.
Can you expand on the defining of the exact location on the image ? How do you go about it exactly ?
@@lebleulebleu1274 You basically tell ChatGPT how it should arrange the stuff you described. Then you need to ask it to transform the picture description into a picture generating code like CSS uses for Website layout thats readable by AI. Define the Character Limit of your Chatprompt and here you go. I only tinker around with the free version so iam limited to 260 Characters. But with longer allowed input you can fit into it a exact description of the objects and how they should be placed. When you have long enough input allowed you could also exactly define how the objects interact with each other. Basically painting with words and code ^^.
I tried feeding these steps into the Version 3 and it's the difference is pretty big. It's wild so much better version 4 is at understanding the information getting fed into it.
Good video my dude
I’m waiting for the day when they combine midjourney and gpt so that the AI can recognize what the results are and you can simply tell it to make adjustments as you would tell an artist to do. Imagine simply typing “I wanted the sunset to be a bit more orange” and boom it goes ahead and spits out your image with a more orange sunset but without changing anything else.
on its way ;)
Yeah, this is what I'm hoping for. Sick of messing with complex software like Photoshop or Affinity, but if photo editing is as easy as 'Expose this area a bit a more, and sharpen this tree" - without having to learn about masks and layers and sublayer menus of sub-menus under menus, then I'm in!
@@bigglyguy8429 You want AI to do everything for you lol. Lazy already.
@@UltimateQball That's what it'll probably end up doing for the most part, though. lol
@@XcaliburZeRo Being submissive to it means that alot of humans are weak and has admitted defeat lol. How about we balance our lives with AI. People are truly living life with fear, if they're already allowing AI to take over their entire life.
In the future, ChatGPT will be customizable, allowing users to interact with the program through speech. In addition, users will be able to choose different "Skins" to personalize the ChatGPT interface. These packages can be purchased at varying prices, ranging from basic secretary models to advanced models like the T800 from Terminator. Those who do not purchase packages will have access to text models that are more advanced than current ones, but still limited in terms of available options. Another detail is that the registration process for access to ChatGPT will be much stricter in order to control all the content that is being created.
y- yeah
MJ is really good but my favourite text to image generator is Blue Willow. I produced some incredible artwork with Blue Willow. Blue Willow recently launched V2 with a brand new model upgrade. More importantly it's completely free
I am getting AMAZING images from Midjourney with all these instructions. Thank you!
could you please publish on your website all your prompts of this video and for other videos if you want?
Yes :) will make a post on this tomorrow
Thank you for this. I am an elementary STEAM teacher and you helping me, and my students, understand the power of technology, and the appreciation of art.
Holy crap. That is really good. Not a doubt in my mind that companies would pay a decent amount to have images like these created. The Viking girl screaming and the close up of the bees were both absolutely phenomenal.
Why would they pay -anything-? They can just type it themselves. Any kid could. Prompter will not be a viable career path if that's what you imagine.
@@kimsyberg720 Yep. Artists won't sell to companies or consumers. The consumer will make their own content. Or tell the AI to search trending prompts. We already let companies/the masses tell us what to like and what to buy. Don't give up your day jobs fellow digital artists.
I agree 100%, I believe a new job sector called *prompt engineers* will pop up.
@@TheChromePoet Seriously anyone with an IQ above 110 should be able to do prompts easily, with just a bit of practice.. it won't be a career.. especially since there's already soooo many people doing this right now
@@tissoeh the majority is not able to find information with Google. So there will be a lot differences in mastering prompts and Just typing in "Picture of Something".
I used the same prompts as you but chat GPT did not give me the prompt in the right format, and did not fully understand how Midjourney works, idk why
Love Love Lovely!! Wish you give an attempt for GPT-4 + Midjourney V5 for a consistent character generation for story books .. Same character with different poses and at multiple location and with different facial expressions ... Is that possible to achieve now?!
Thnx for tuning in :) we explore that this weekend :)
Yeah was also waiting for this!
I'd like to see you put in "dumb prompts" like just going to Midjourney and ask it to draw a plate of pasta to compare the two. Amazing stuff!
Would have been nice to compare these images with those you get by crafting some "naive" prompts yourself rather than using GPT4. But pretty neat process, "teaching" ChatGPT things this way seems really powerful.
Exactly, this is all bs, I get the exact same results with 4-5 word prompts…
Prepping Chat GPT feeding it relevant training data in a single chat (copy and paste), and telling it to store it for research data to use in further responses to prompts, allows GPT Chat to be tailored to outpot exactly how you want
They explicitly say that in MJ5 you don't need to work it up with all this fluff, just RTFM and tell Midjourney what you want to generate
I asked GPT a basic question about a prompt to feed Midjourney and it already added to my very basic description.
"For example, you can try “Generate an image of an old brass microscope similar to Leitz, Zeiss and Watson microscopes of the late 1890s”. You can also add more details such as “The microscope should have a brass body with a black japanned base” or “The microscope should have a rotating stage with two stage clips” to make the prompt more specific".
I was actually impressed that it was aware of Japanned bases, rotating stages with slide clips, which were common back then.
Japanese ?
I wouldn't call this photography. Its just art that looks like photography, but I believe AI is going to hit photography hard, and there are tools to replace a lot of these things and open up different job avenues.
It's just photorealistic illustration, same as we've had for a few years now, just faster and easier...and a bit better. I've seen work like this for at least the past 10 years over on Deviant Art. But once it's so advanced and easy that any soccer mom can enter prompts as quickly as writing an email, and get a photorealistic result in a couple of minutes, then there will no longer be a career path for anyone who knows how to use this...except perhaps at the very highest levels. From a commercial standpoint, I think it will be more valuable as a set of advanced augmentation tools for editing actual photographs. But, yes, product photography and low to-mid-level portrait photography is gonna take a big hit for sure.
@@RobertFalconer1967 I agree with everything you said but....there is no but. It's going to be scary . Many will have to learn new skills.
@@AndyHTu Oh definitely. But that doesn't *necessarily* include photographers. Event, scientific, documentary, wedding and photojournalism are all safe. Anything that requires a real documentarian to capture real events in the real world in real time will always be needed. And ironically, the more fake images are propagated across the Internet-created entirely behind a keyboard-the more "authenticity" may well become the new requisite currency for imagery in many fields.
@@RobertFalconer1967 Man I feel like I'm talking to myself today. Everything you just said is what I was talking about earlier on Twitter in some of the Midjourney people's tweets. Its pretty incredible! But I'll add this as I've mentioned this in the past before. Its going to be huge for the fashion industry! I'm already working on concepts for it and the types of jobs you can get it are like you mentioned before, are mainly going to be catered to Creative Director roles. I think this is going to be the case for man creative industry and programmers and writers now unfortunately. I'm not sure where this leaves the lower end jobs, but fine tuning models will be pretty important and hiring artists might still be important to inject more specialized data. This part I am just thinking out loud as I don't even know if it will be necessarily in the future.
@@AndyHTu "...creative industry and programmers and writers now unfortunately."
Low-end writers like copyeditors? Yes. Creative writers? No. It's still terrible at that. What it churns out are perfectly constructed sentences with awkward, if any, narrative, and zero nuance or human subtly. It cannot read your mind and know what you're thinking about the sort of message you want to convey to your audience in an article. And writing at the high end level-novels, screenwriting, etc-have nothing to worry about until A.G.I. arrives. But at that point we're gonna have far bigger problems to worry about.
Also, the more fakery that is propagated, the more knock-on problems are going to emerge in our society. And quickly. I wouldn't be surprised to see an entire industry emerge rapidly of new A.I.-based programs that run any creative content suspected of having been generated by A.I. through their systems to determine human authenticity and accuracy. New cyber "watermarking" systems may also have to be implemented across the imaging field if said images are to be used editorially or commercially.
I’m very experienced with V5. It’s actually detrimental to input such detailed prompts currently. Keep them short and straight to the point.
Does it make sense to call this photography? That's specifically the recording of light via an image sensor.
This is just mashine algorithm, noting more. . it will make people dump and useless . It is mind-blowing that here in the comments there are so many exited people. That is insane😂
I guess this is a similar question as to "Is chess really a sport?" You don't do physical exercises, you just sit there, maybe walk to another table.
I heard it is was defined as a sport before there was a clear regulation as to what makes a sport in of itself.
So if we define photography as only the act of taking a photo, then no, this wouldn't be a photography. But on the other hand photography has become much more than that, since it includes now heavy digital editing. But since this it technically "only" editing without any photographed input (if you discount the photos this AI was trained with), you could call it "photorealistic AI generated digital art.
btw instead of copy pasting from a webpage you can actually just give GPT a url and it will read it
So essentially they
are close with replacing all photoshop/ photography creatives , it's kind of sad for people who trained themselves for years in in this
Tough luck
Great tutorial on priming MidJourney. I'm looking forward to two updates in MidJourney. 1.) Ability to get hands generated correctly 2.) Ability to generate text / text on graphics
these text to image ai systems will use gpt4 to filter unusual pictures like weird fingers. Because gpt4 has eyes and detects unusual details. That's gonna be how gpt4 will actually revolutionize ai art
Yeah but tell me about the porn?
@@bigglyguy8429 bro what 💀
Great example of priming GPT-4. Have you experienced any problems yet with the length of your conversation? Very long prompts might exceed the maximum sequence length, and for very long conversations, some of the conversation history might fall out of scope, i.e. GPT-4 will not "remember". Have you experienced any of that? Do you have access to the 32k token model, or are you using the default 8k token model?
AI went from slowly crawling to sprinting full speed, and it`s good to see people are on it, finding new ways of improving it`s usability. Awesome stuff.
Super, where did you find the text of introduction on MJ to teach to GPT ? Thanks
It's not photography, it's still digital art, still amazing tho :)
it's not art either, but it is incredible.
That's Ragnar Lothbrok played by Travis Fimmel from the show Vikings at minute 8:46. Still producing copyrighted content I see. Stealing work from other artists is not ok.
You all know the consequences when an artist was rejected from an art school in the 1930s. Now millions of artists are going to be mugged. Just sayin to not be surprised.
I’ve been following this stuff for 8 years now. It’s definitely wild how fast it pushed in the past 6 months. But I think it’s just going to allow artists a better tool. How art is profitable is constantly changing. Always has been. Artists are the ones using this to do better work easier. This is breaking down technical barriers to make great work.
💀
hungry?
@@StefanMaierMediengestaltung grab a Snickers!
@@Cragdognamedbear it'll allow artists a better tool for a very short window before making commercial art obsolete. That's an inarguable fact.
Fine art won't become obsolete, but it will become more competitive and an even less viable career than it already is.
I'm a 2nd Grade teacher who is going all-in with AI in education - these videos are PHENOMENAL
I won't be one of the educators who perish into antiquity - this is a revolution and there are some teachers who are leaning into this movement 👍🙌
Cool! How do you apply AI to your lessons?
@@mauricemakesmovies so far, it's been incredible with eliminating redundant tasks like creating word problems or journal prompts that are truly personalized to my students' interests. They love Minecraft and know way more about it than I do - I throw my 2nd Gr math standards into GPT-4 and it creates diverse, differentiated, and contextualized math about Minecraft. My students are engaged without any effort and that is priceless as a teacher. I don't have to convince them to learn when they are interested in the material on a personal level. I also find that I learn so much from the intuitive nature of GPT-4, for example I have learned so much about Minecraft from the way the characters or materials are included in the word problems. It amplifies creativity.
I also find that when I ask GPT to connect two concepts that don't seem alike, it unlocks connections that I would have NEVER found on my own. I asked about how erosion and 2D shapes are similar and it suggested that I do activities with wood and sandpaper, turning shapes into other shapes thru weathering/sanding 🤯
I have noticed that GPT-4 isn't some revolutionary teacher that produces novel ideas all the time, but it provides many excellent examples that can be used as a springboard for truly creative educators.
In teaching, there are Sustainers, Implementers, Innovators, and Visionaries. Sustainers maintain the traditional system. Implementers try out new ideas that are passed down to them thru traditional channels. Innovators combine best practices to create new practices. Visionaries are the teachers who aren't afraid to say "Wait, Stop! This entire thing is broken beyond repair and we need to change everything. Let me show you how". The Visionaries are the ones riding this initial wave of AI
I was impressed! Viking women warriors was amazing. And pasta V2 is incredible. Flakes of cheese flying in the air is 👌 Add some dynamic to photo
Even food photography is not going away! Because if I need to do a new shot of a dish for a menu, I would still require a photographer who can take a photos of dishes, instead of generating some random food. Same goes with wedding photography or wild nature photography, or even the green comet (which recently was shoot on cameras all over the world) ☄️ Maybe for a mock up of a website it will work if you need a photos of random people or random food, instead of using unsplash like I’m doing now. 😅
lmao keep huffing that sweet copium, its going to keep your job for sure mate.
"Instead of generating some random food"
But why would you want to do that? You could take a photo of a dish yourself and midjourney V5 can take images as input such that you can create better versions (professional looking ones). Doesn't this approach eliminate the need for a photographer?
Can AI wander around a wedding and take documentary photographs of real moments? No. It just scrapes existing images and normies go mental.
people have been making real websites with it for months. not mock up stuff. I've done it. no photographers hired...
Food photography is going to die. No food IRL ever looks like the pictures taken for ads. So, no one really cares any more. Sorry.
As for Wedding photography, yah that is save because people want real images of themselves and guests at weddings.
Nature photography.... eeeeh, sadly it's now a dead industry. But, it shouldn't be.
It would have been extremely helpful if you had linked where you found the MidJourney Text that you Copy/Pasted or at least shown the page in the video I've been searching now for hours and cant find any of it.
It's brilliant, but I started playing with the programe a couple of days ago. By weird coincidence I wrote a prompt of close up of Viking shieldmaiden's face in battle rage and it produced something similar. Very similar....same lady with just slight variation.
This was clearly after you. It remembered yours and produced a close variation for me.
Mine was after 11pm on Saturday 26th March.
I'm thinking this isn't as "original" as we are led to believe.
Happy to share my images with you if you message me.
I have no axe to grind here. Just curious about the tech.
One my concern with making high skill level art easy through AI is that it gonna turn it generic and not interesting as a result of there being too much of it and it will just feel like soulless stock photo compilation
There is even a difference in perception of a highly detailed art piece that you know was made by hand "that person is skilled" and knowing it was generated by ai in a click, making that art not as impressive and it wont produce any good impression apart from yea it looks good.
We are so close to making everything obsolete. The interior design thing for example. You let the computer design your interior, then it sends the information to 3D printers to actually produce the stuff it designed, their computers send you a quote, which you pay online, then a robot loads the 3D printers with the needed printing materials, a packing robot loads it into an automated truck, which delivers it to your doorstep where you've hired a couple carrying robots who put the stuff exactly into the room as it was planned. The only time a human is used in this whole process is when you type the prompt to get an interior design for your room.
that's too funny lol. like typing random keywords to find stolen images lost in the latent space of the model wasn't easy/effortless enough. now lazy, talentless techbros are stacking algorithms to compensate for their lack of creativity and technical skills as well.
what a time to be alive.
So basically you don’t know how MidJourney works
Outstanding! Just signed up for your newsletter! i've subscribed to GPT, not sure if I'm on GPT-4 or not lol....and I'm trying to figure out Midjourney! I'm writing a book and I want to use MJ to craft images that I can't find anywhere else....thank you for this great video! Subscribed and looking forward to additional content!
I followed your instruction and gave chatGPT the 5th prompt, then it automatically made a 6th prompt on its own. blown away.
10:58 LoL '...something there...', yeah... It's just the copyright watermark. Let's ignore that. #WelcometotheFuture
AI integration is the evolution and future of digital art, photography, commercial advertising, gaming, animation and story telling.
AI "editing" in photography is more akin to the evolution from film to digital photography, and the introduction of digital editing software. The food photograph glow-up is a fantastic example, and is what I'd assume is the future of AI in photography looks like, in the immediate future.
6:50
There are actually subtle tricks to make food look better than it really is like literally using glue to make it look like Mozzarella, which is a trick that makes it "look good", but it wastes perfectly good food.
kriiis you are insaaaaane , i cant imagine that i was able to get that quality of photo from midjourny !!
thank you maaaan from bottom of my heart !!!!
This is freaking genious man! Subscribed :). Excited for your newsletter
Storyboarding for films just got a 10000% upgrade
This has truly blown me away. I trained my chat-gpt with your prompts and the pictures i'm getting off midjourney now are now next level. Thanks for sharing :-)
how?? I've done it to the tee twice but no cigar. chat gpt only writes stories for me
@Mighty Marce I'd check how your doing it. I followed his instructions using his script and fed 1 to 10 it into chat gpt v4. It's most definitely working.
今回初めてあなたの動画を見ましたが、Midjourneyに関する詳細情報をGPT4へ教育する方法を丁寧にレクチャーしていただき感謝しています。提供してくれたマニュアルには他の使い方も掲載されているみたいなので、日本語に翻訳しながら活用させていただきます。貴重な情報を本当にありがとう、Bro!!
my chatgpt dosent understand the prompts, perhaps because i use the free chatgpt? while i am instructing him with your prompts after prompts examples (9) i think he forgots them and say "Hello! How can I assist you today?", and then he dosent uses the promt format, he uses natural language to describe the scene.
ok, I've tried to make this work in GPT-3 and it returns the prompt description out of order, you need to type:
"Don't forget to condensate all those info in a prompt like this: {eg some prompt of the PDF file}
Another great content, m8.
Great tip :) thnx for tuning in :)
I tried chatgpt yesterday for coding website.
Everytime I ask to make small change, like alignment, it would spit out a completely different code with styles etc.
Not only that,but it cannot remember a thing I say.
I would give it a piece of code to remember as code1 and then another piece as code1, then ask to combine the two and remember it as code3.
It cannot do that.
It gives me incorrect number of lines of codes I told it to remember.
When it combined the code, it was a random mess.
I'm completely blown away at the quality!
I think I heard you mention that I can use an original photo and then have a I clean it up? Do you know if it's possible to take a background photo and have a i add a different picture or combined two photos. For example a picture of a restaurant without any people and add people sitting at the tables?
10:58 confirms my worries a bit, as all the images in this video just look like variations on pre-existing images
The compounding effects are amazing.
You are a great help. I am learning as much as possible from you and others. I am now learning everything you have done in this video.
You sound like an AI yourself, Sheezus.
This is very important as not many people have the knowledge to be this detailed in explaining the prompts.
Soon it wil work without priming.
Amazing video! I am into AI so your channel is amazing!
Thanx Alen:)
1. Would you explain insertion of a photo into Midjourney and how it creates or modifies the image? Secondly, What if we insert a photo of the last century , (how do we insert and then aske Midjourney to improve the image quality without changing the content or background? Any examples of the second?
the 3 feet lady at 10:47 is Virginia from Sons of the Forest!
Fantastic job, Just tried it out.
Is it possible you could make alike training for KaiberAi?
Nice work here. Joining the Newsletter. 🙌
You should tell it to output only the prompt and inside a code block so you can copy it in one click
They will never truely completely replace ALL photographers, but if you are a low-mid range photographer that "copies" other styles, then yes, obvously you're out of business. But all documentation photographers (weddings, news, events, specific portraits..) those will still be needed. Also high production advertisement will always be a thing. I see this replacing Stock websites VERY fast, those photographers are completely out of business in the next 5 Years. Adobe has already stuff going on and pretty sure IStock will follow.
This is working incredible, you can also teach it to write the prompts in a code block, so you can copy it faster!
i know very little about technology and i'm bot fascinated and scared seeing all this updates, the response is very human.
that "yes i understand how midjourney works" sent a little chill down my spine
Hehe, after giving it some of your prompt examples, ChatGPT came back with this:
Sure, I'd be happy to help with that! Here's a poem based on Prompt 4:
The snow-laden slopes stretch wide and far,
A winter wonderland beneath the sky so blue.
Skiers and snowboarders carve their way,
Down the mountainside, a thrilling display.
The Japanese architecture stands tall and proud,
A nod to tradition amidst the powdery snow.
Evergreens laden with snow, a peaceful crowd,
Nature's beauty on full display, a captivating show.
The camera captures it all, with precision and skill,
A Nikon D850 and Nikkor 24-70mm lens to thrill.
The aperture is set to f/8, the ISO to 200,
A shutter speed of 1/500 sec, the perfect blend.
The result is a stunning photograph, a moment in time,
A breathtaking winter day at a Japanese ski resort divine.
I confused star wars for star trek when you said star wars, so I looked at the pictures, and thought "that's not star-trek inspired, it's star wars, then realized I had heard wrong, so the design is so strongly star wars it went against my misheard bias.
14:05 Chip Foose you have been fired....LOL
I have a friend that does food photography for a living, the details in the photo that you prompted with the pasta is on par if not better than what he does. Now imagine midjourney in 5 years, it will probably be able to create movies on his own, you just write the story with him.
The issue I'm having with midjourney is when an image is amazing and spot on, but there is slightly small issue with fingers/eyes/unrealistic imperfections. There is no way to tell it "I want this, but fix this area". You need to regenerate and get almost perfect images with different issues. The problem is that if 2% imperfection in the area of the eyes can make the image totally unusable. If there is a way to overcome this issue, I think it will make a great video (it's a common issue and no one posted a solution on youtube yet)
photoshop
Hey, for my graphic novel I'm trying to build micro photography....
thank you so much for all of the info! heads up!
As a marketer and brand guy the opportunities are endless really speechless
Oh wow thank you Kris. Looking forward to having a look at the prompts in your newsletter.
It's getting closer... but... I've come to believe that a huge issue, currently (bring on v6), is MidJourney trying so hard to treat a single composition in it's totality and tie all composition aspects back to the biases it has "learned". Sure you can dive into weighting... but the next step for MidJourney's evolution could possibly be introducing prompts like - - face, - - left foot, - - right foot, - - torso, - - horns, - -background, - - foreground... so that users can then give MidJourney a little more help deciding what to do and a little more power to treat different parts of the composition independently rather than as a whole... and THEN try to rationalise it all into one composition (or not... if the user really really really WANTS a very disparate composition they should be able to indicate such with a - - frankenstein value (for want of a better term). Obviously this applies mainly to compositions featuring humanoids. Those using MDJ to make websites would want a similar ability to compartmentalise.... but with terms like - - header, - - navbar, - - menu, - - slideshow, - - footer, - - sidebar. It is possibly time MDJ learned how to separate so that, for example, 1 - the horns I requested on my Tiefling don't just look like horn shaped hair (because it was too close to the hair), 2 - the black leather magic tome I requested in his LEFT hand isn't set on fire simply because I also requested a Fireball in the RIGHT hand... and 3 - the little girl can go barefoot. Heck, then maybe even take it further and allow - - cellref C5 where A to K is the distance along the horizontal and 0 to 10 is along the vertical... so I can request - - cellref C2 barefoot - - cellref C8 barefoot.
I would love to see a comparison of inputting the gpt prompts to the simple original prompts
The exact same images lol
@@GayClone Not at all. Simple prompts generally do not provide a lot of unique detail with mid journey and you can get some strange/uncanny images very easily without a good prompt. I take it you haven't actually used midjourney before?
Not going to lie; this is the most brilliant lead magnet I've seen for getting me to subscribe to your email list. Done!
If we hook up ChatGPT4, MidJourney and the Star Trek replicator, all the meals will be amazing.
In your first example, you accidently didn't copy the whole prompt. You missed the --v 5 parameter which triggers using version 5 of midjourney.. thus the hand issues
Why when I follow your steps GPT generates just a general description and not a MJ prompt? this is frustrating
THANK YOU SO MUCH FOR THIS. THIS IS GAME CHANGING FOR ME.
First video I've ever seen from you, absolutely brilliant, I subscribed quickly!
Most excellent Thank you very much. I find that after a little while it starts to deviate away from the original prompts. but this is easy enough to update.