Chapters:
0:00 An Idiot Talks About Making a Book
2:53 Step 1 Midjourney Renders
4:04 Step 2 Script Generation via Inferkit
5:47 Step 3 Layout The Panels in Comic Life 3
13:21 Step 4 Design A Cover and Publish
16:02 An Idiot Shares His Recap
Midjourney is a pretty incredible AI tool for creating images you can use to create your own graphic novel in less than 24 hours.
Check out the time stamps to see a timelapse of each step and my tips on the process.
Using artificial intelligence to create full comic books isn’t as far fetched as it sounds. This video lays out a high level walkthrough of how I can create a full graphic novel in a single day using AI (Midjourney and Inferkit) along with Comic Life and Affinity Designer.
Outside of Midjourney I also use these apps/sites but you could do the same thing in GIMP or any other free design software.
Inferkit
https://inferkit.com/
Comic Life 3
https://apps.apple.com/us/app/comic-life-3/id891378056
Affinity Suite
https://affinity.serif.com/en-us/
![Full Beginners Guide to AI Art – [MidJourney AI Tutorial]](https://i0.wp.com/aiartvideotutorials.com/wp-content/uploads/2022/11/563/hqdefault.jpg?resize=350%2C200&ssl=1)
I was half excited to see what you would make, but the results are less than satisfactory.
Wow. After 30 years in animation and comics… I am working way too hard. AI is taking over the world and making me obsolete. lol. Good times. Great video.
We’re at a historic period in world history, and most people don’t even realize it! The entire field of graphic design and maybe writing are about to become completely automated! What that’s going to do to our will to create I don’t know!
Another idea I was thinking of is to do a graphic novel video, say on YouTube, using Midjourney or one of the other art generators using a public domain story.
how do you create the images to have the same boy or same dog on different images.
Fantastic! I am going to watch this over and over and get this down.
very informative video, i'm gonna try doing the same thing using Stable diffusion, and also, what music are you using in the video at 5:30?
This is brilliant!
“We’ll we’re not at that point where it’ll replace artists yet” why is that an argument for it? Just gonna wait around and feed the beast that’ll grow to consume you because it’s in an infantile state?
“It’s not Salvador Dalí or Picasso” neither are most artists that make a living doing it. No shit it won’t replace prodigies that we celebrate above others. That’s less than 10% of all artists ever.
But whatever right? The comfort and convenience that technology provides is worth the unintended consequences. Instant gratification is a good thing for our brains, I can’t wait for a majority of the population to be iPad babies with 3 second attention spans.
Reply to argue if you want but you’re wrong and I don’t care about what you have to say. I will not be convinced and neither will you. Only time will tell
Affinity is great. I only use Adobe when a client makes me use it . I started using Adobe tools in the early 90’s starting with all the 1.0 versions. Im done with Adobe and their greed.
that's a great concept .. so many possibilities
but I ask myself what kind of person would read your depressing dog comic lol
super info thx man
So you don’t do the art or the writing? When people ask what you do for a living, what do you — never mind.
In reality …don’t do it in a day… you won’t get the best results. The ver3 graphic novels people have done so far will already look very dated very quickly… even now they look dated given the new algorithms on MJ. At least we have test and testp now, and you can use ver3 remastering as well which uses test which can be very useful. Using those also takes much longer to render.
But have some integrity and don’t want to spend 5 minutes generating images that are merely passable. You’ll end up with shit detail, wonky AI weirdness especially if you need faces to look good. You should also care about whether it actually is artistically working for your actual story. The least you could do with MJ is to spend more than a careless 5 minutes making your images for something specific like this. Otherwise you’re just throwing out shit just to be fast. You should care about the images like you want others to care. Unless you want it to have 3 eyes and a weird body don’t just accept it. Work on it so you’re proud of the end product rather than just trying to get people to buy your shit 5 minutes work. It will also get way more longevity and also you won’t feel embarrassed about telling people how long it took you, people won’t be impressed if they found out none of the images matter and you didn’t care.
Also a good point about upscaling is that they don’t just upscale it’s quality and resolution it literally gives you more/different details. There’s Light (uplight) Detailed Upsxale and Upscale to Max and the newer one Upbeta. None of them are the best, they’re all different. They can substantially change your image! And not always in a good way. Use Topaz Labs product or something similar to increase resolution if none of them are good enough for you.
Here’s a good tip I’m using now. Make something in ver 3 then remaster it after upscaling then do Max and Beta upscale and remaster them as well. This way you can see how it looks in all of them. It can be a HUUUGGGE difference, and then when you get your remastered ver3 image you can also check all 3 upscales to check the difference as well. ALSO something you should know is that everytime you upscale you get a DIFFERENT version of those extra fine details it adds, so if something isn’t quite right try upscaling again. If you’re at the end of the remastering process you can even go back to the v3 image and start the upscale>remaster>upscale process all over again.
Also when you find a good image use the same seed by reacting to it with a
and the MJ bot will send you a message with the seed number. Put that seed number in at the end of your prompts like —seed 123346 etc, then it will help maintain some consistency. If it needs or be said… figure out your seed you want to use and all the general style prompts first if you want a consistent theme. And yea, that will take some time but it SHOULD be fun to figure out.
There’s also things you can do like taking a test or testp render and putting it back into ver3 as an image prompt. Weight it high (or low if that’s what’s appropriate) which is 10000 ie. —iw 10000 and make sure your prompt describes what you want the differences to be or if you want to see what happens if you try and recreate (it won’t be the same) describe exactly what’s in the image, for that you need to take into account the different way ver3 processes prompts compared with the much more limited and less create testp. This means you can go through the ver3 upscale remaster upscale process and then repeat the process again with that image. It’s worth saying that some stuff remastered very well and some stuff doesn’t. Some stuff remasters so well it can make a very wonky ver 3 image to something great. Like, it made that weird smudge a child, or a bookcase, or whatever. Great!! Ver 3 can give you very different images so it’s worth experimenting rather than just going to test
And then you can always use apps to fix things further, obviously photoshop but other much cheaper apps like RetouchAI which can help you fix a lot of problems. You can also use an app like Remini to improve eyes to some degree, although I’d recommend paying for it because having 3 algorithms so much better than the first free one
Saw your post on midjourney, thanks for making this!
Sir, you can’t justify using a program and AI to steal other artist’s work and create something that you didn’t even really come up with. That’s ridiculous. You’re mad cause it’s true. Why don’t you take the time to learn how to actually create and draw comics? I’m 28 and make my own comics and do studies. AI generated art is soulless af. And the fact you don’t see an issue in using someone else’s art to make a profit is very telling. The cognitive dissonance is real.
What's a good Comic App for the PC? I don't have a tablet. I do have an old Macbook Pro…does that app exist for Macbook Pros too or just the tablet?
AI is a muse that extolls too much for me personally. Good for you, but for those saying they wasted their time- art should be a little painful.
Very interesting video, I enjoyed. Sidenote, you look like Dr. Cox.
I only have a question: did you enjoy making this? Did it give you a sense of purpose and meaning? Of true achievement? I love drawing and I don't understand why someone would treat the whole process as a chore to get rid of in a day, besides of course speeding up some tedious parts. I would like to understand the mindset behind this because it escapes me. On average art is not especially profitable so why a person would be fulfilled by claiming as theirs art they didn't even make themselves besides the vision behind their prompts? I would understand if at least the text was written by us then we seeked a way to make our ideas visible in case we didn't like drawing or we couldn't. But having even the story and dialogue generated by artificial intelligence? What's the point? It's not a rhetorical question, it's that I genuinely don't understand.
First of all refreshing to see a tutorial that is actually helpful. Is there any way for me to incorporate my original character into Midjourney so I can generate scenes through it's program?
lol
AI ain't shit
Thankyou for sharing, great inspiration and helpful tips, greatly appreciated.
Thanks man I'm a 46 year old digital artist and you just set my path on fire I've used programs for years published 47 books the hard way you just gave me my creativity back what you're doing and the simplicity is ingenious screw the negative people you are ahead of your time thanks so much for giving me my drive and passion for art back hope to see more of your work and find a way to send you something when I've completed one
Whenever I use GMS, it softs notNice tutorialng like yours, even though it has the sa settings… It's just a buzzing noise. Weird.
I don't think everyone is obsolete. What always happens with time saving tools is the studios expect you to do more. We're not yet to the point where an exec pushes a button and a GOOD comic/movie comes out the other side.
There's no copyright protection. No major publisher will go near AI comic books.
just got your video suggested by YT. Great content. i messed around with midjourney. great fun, really addictive.
but what a great idea to make a book with it. never came to my mind. 
i subbed and will watch more of your content. really appreciated. greetings from turkey. keep up the good work.
Fuck all that waiting on a human artist when I can holla at mid journey. That’s like riding a horse across country when I can catch a plane cheaper , faster
. Come on now we love horses but the time is up for em to be used as a main line. As a consumer that is just buying the art, trust me we are just going for the the best picture. Price and the creator is not a factor depending on how beautiful the art is. If I was a artist I would be learning how to incorporate this in my work or to help me create better work. If not your gonna be left behind unless your an exceptional artist that can’t be copied by a computer.
Where do we buy your books?
Would be fun gifts if you decide to offer print versions (I only found Kindle version of Meating People)
incredible breakdown! I really want to give it a try
Imagine Andy Warhol finding this during his lifetime. That would be crazy cool!
Ive got adhd too and can feel how this could help my creativity! I need to get thing done fast or my interest is out the window.
Go to the high seas! Piratery is FREEDOM
I'm 30 years old now.
I'm very impressed with Midjourney, I subscribed the $30 plan, and I finally got able to work on making comics of screenplays which I wrote in 2006 and could never produce a comic or record a movie about that until now.
I'm also working on bringing to life even characters that I idealized in 1998 when I was a kid.
Midjourney practically changed my life.
Phil, this was so awesome! First run took me about 48 hours and boy am I stoked. Great job and awesome video man. Only took 15 years for me to get on Kindle. My graphic novel, NEONSOMNIA comes out soon!
That's that's Incredible.
If the ''creator'' doesn't write the script or draw the story, they aren't really creating anything they're just assembling machine-generated, synthetic content and if we care at all about art they should have to credit this rather than be able to put their own name to a book – eg the author and illustrator of the book is Midlife rather than themselves, and they should be credited as the 'assembler'. To my mind, just because the artist is a machine and not a biological human shouldn't make any difference. It's down to who or what is doing the work and this should have to be made clear. If that is we care about art, culture and artists. If a guy didn't write or draw something he shouldn't be able to put his name to the book, end of. It's a book written and drawn by 'Midlife' or whatever software programs they used. If this guy were to just rip off an artist (or keep him chained up in the basement producing material) and call it his own work there would be problems, so why is it any different if he's getting a machine to do the work and then taking the credit ? Rights for AI anyone ?
Making the art using AI is one thing, but how creatively bankrupt does one have to be to use a text generator? Is it really that difficult to write your own story without AI barfing out text for you?
TL;DR of all comments here:
1. You can't replace hand made human artwork.
2. True, but we artists who barely survive on custom art commissions are almost absolutely f***ed.
where did you print the novel for 10 bucks?
Great channel
Thx for the tutorial
Ha! I thought I was the first one to think of this!
Can’t wait to see what you did
Excellent video, thanks for the great content and your delivery style.
wow this comic looks like shit XD, maybe in the future AI would be better but right now it's far away from replacing real jobs like AI art looks pleasing, but once you open the photo it doesn't look understandable, and btw drawing human emotion and poses is 10000 times harder than drawing nature…etc anyway let's see what will happen in the future but right now it's still got a long way to go
btw if you are an artist don't be afraid even in the future if you somehow by a miracle got replaced, your skills could translate easily to other fields like animation, motion graphics, design, 3d modeling, 3d anmation….etc
but if an AI could replace all of this then we have a much much larger problem than you having a job or not XD
I can’t watch this. Just stopping by to say I’m depressed
Really nice, it has a consistent avant garde feel. Would love to know the input for each images.
Como hiciste para que todas las imágenes tengan el mismo estilo antiguo de tinta
For folks wondering how to maintain consistency across image generations, or how to generally transform any existing image be describing that transformation with words, check out this approach: https://www.youtube.com/watch?v=_CtguxhezlE
I will never use this junk. Ever. And for that, my work will be infinitely more valuable than anything made by a toaster. Toaster art! Lol
I write and illustrate my own Comic Books and Graphic Novels and rely a great deal on Digital Art. Mostly I use programs like Photoshop because they really reduce the time factor, but using Midjourney is in no way a threat to graphic Artists or illustrators and the original content we create.
Truth be told we should explore these software avenues to see what we can develop from it or use to enhance the original work that we have in mind. Don't limit yourself people.
P.S.
Excellent video!
You're a very clever man, Phil.
Next up, a tutorial of how to make an AI to cosume media for you. So you can spend your time thinking on how an AI could simplify the current task you are working on.
I've published two comics so far using MidJourney, Lungflower and The Eldridge. I generate images in MidJourney and then tweak and assemble them in Photoshop. The technology still has a long way to go, and it's a challenge, but it's very possible to do.
Yuuuuuuuuuuuuuuuuuuuuuuuuuuuck
How to FAKE a PIECE OF SHIT in one day.
Seriously, are people's egos that fgdged up that they are pretending that literally doing NOTHING is a skill and that they are doing it themselves?
W O W!
I'm actually done with you clowns.
This is actually brilliant! I will definitely have to do some testing myself with your tutorial
how do you keep the looks of the characters consistent using midjourney?
How did you create enough coherent images in midjourney to have enough choice to fit the dialogue?
How to make yourself completely redundant while churning out mindless drivel in ONE day. Truly inspiring stuff.
Can you change the art style? Like making it more realistic or stylish. Cause it seems cheaper than an artist.
who is the author and who is the artist on this book? did you write your own name? what's the legality of this?
No thanks. I'd rather use art.
I don't like this at all. The AI art is very close to Basil Wolveton… who can't give his permission or make any royalties of this work. This basically feels like a demostration of art theft. He took years to work on his style, and it is getting swiped in days. Kind of disgusting.
Brilliant tutorial – thanks so much for sharing your workflow. As an artist I've really embraced AI – it's just another tool that either helps fulfil my expression or provide inspiration for pieces I might want to create using more traditional methods. I really hope both Midjourney and Stable Diffusion start providing us tool that will allow better continuity between images – kind of the Holy Grail for AI Graphic Novels. Would be awesome if we could train AI based on AI created images. Textual-inversion is a good starting point and would be great to see this feature on Midjourney!
not sure i get the point of this but for those who do…great tutorial!
Amazing
Phil great video! Wondering if you ever had this problem where you started your project on a script in the comic life app. Am I able to transfer it over to the comic portion?
Interesting video, and I’m trying to wrap my head around how this can be useful to an actual traditional artist as opposed to some talentless yob who wants to churn out impressive looking drivel for social media likes.
Hard to be happy about AI rendered art encroaching on media that used to take talent to make by hand, but I’m trying not to be too judgmental or negative yet. Thanks for the video.
I do find the robotic generic electronic background music is kinda a perfect match for a “creative” endeavor like this.
Do you have link to see your book ?
Midjourney and Tik Tok will end of the world as we know it… and it all began with Napster
Artists will never become obsolete. This tech may be used by corporations someday to create fake commercial “art products” but people who actually like art won’t give it the time of day. Real art-loving people want the authenticity of work that’s been created by a real human, with their own real human ideas and a real human personal art style. That’s a big part of the appeal for most of us. This tech may make the commercial world value artistic skill less, but that’s nothing new when you look at the past 200 years of technological innovation in the art world. Photography, printing, software, etc.
This is a freaking amazing tutorial, I love this idea so much, thanks for the inspiration.
¿And is there a link where I can read the comic? Wasil Bolverton's Barfield the Dog
It's not a graphic novel. It's an artificial graphic novel.
This changes everything. I’ve been spending nights and nights perfecting old scripts, stories, and producing them into storyboards and more. I highly recommend you subscribing for the premium plan and use this for your passion. It will change your life!
Genio! Gracias por compartir todo ese conocimiento. Sería extraordinario poder disfrutar con buenos subtitulados. Necesitamos que YouTube tenga una buena Inteligencia Artificial para traducciones en tiempo real, je.
The market will be flooded with mediocre, meaningless work from people who feel entitled to success and are perfectly fine with exploiting others. Now they don't have to underpay an artist from a third world country! Instead they will be owned by the owners of the new AI tools. Only a handful of these new creators will be relevant, initially as a gimmick promoted by tech companies. Writers who can't write well and rely on pictures to tell their forgettable stories don't understand comics, so keep waiting for that upgrade where the AI does the real creative work for you.
I rushed to midjourney, I typed in a somewhat famous military person from Wikipedia and it had literally 0 images. Not sure what you suggest?
Where can one read the comic "Barfield the dog"?
"I'm just lazy"
sammmme bro lol
This is not sequential art (comic), it's just a succession of illustrations without graphic continuity.
Inconsistent "art" with incoherent "writing", slapped together in a collage. A keyword search does not a comic make. Hardworking artists and writers that can actually tell a story struggle to get by while dreaming of creating something special. Meanwhile… talentless hacks dog barf garbage like this, dragging the industry down with it. I see a use for AI. Creating images of a world for reference, helping come up with something you may not otherwise think of, but come on, this is basically googling images, but not getting in trouble for using them in a "product".
Great video. I'm just wondering if there's a way to maintain consistent characters for the comic in MidJourney? It seems that is the one thing that's missing in creating these visual books is the ability to maintain the characters same look on each page.
This is great. Thank you, man.
Absolutely incredible. Fantastic video.
But..but… How are you keeping the "art style" consistent across your midjourney images? Doesn't it completely change everything when you switch from having a prompt talking about a screaming man to two dogs walking down the street, for example? I'm confused
That’s barely a visual novel though. More like a collage of random images.
I think ai digital works like comic books, wall art and so on should be honest when selling their works as “ai art rendering” because punching words in prompts should not get the same wow value & respect as an actual artist who labors over their art.
nice!!!!
ADHD is a fake diagnosis for business. PEOPLE ARE JUST DIFFERENT.
Very cool
Is there a free alternative to inferkit?
Why are you generating images in public channels and not through the Midjourney bot?
best art sofwares i use are KRITA and FIRE ALPACA and yes I use them more than photoshop. I use them to make full comics and full anime shows together with davinci resolve and fl studio, i make all my shows all by myself, art, music, voices, story writing etc and love and live for it even though i don't make any money yet lol – but yeah I love ai too as EXTRA and thanks for this cool video it's awesome
That moment when you emphasized that the makers of the fonts should get paid for their work and at the same time called attention to the irony that you were saying this in a midjourney video. That was the moment I subscribed
Okay well this looks like absolute garbage lol so I feel a bit better as an actual artist.
I hope you enjoy creating this garbage.
wow