DALLE: AI Made This Thumbnail!
Вставка
- Опубліковано 23 вер 2024
- DALL-E 2 is an AI that can draw anything you ask it for. It's terrifying and amazing at the same time.
Tim vs DALL-E: • Can AI Replace Our Gra...
MKBHD Merch: shop.MKBHD.com
Tech I'm using right now: www.amazon.com...
Playlist of MKBHD Intro music: goo.gl/B3AWV5
~
/ mkbhd
/ mkbhd
/ mkbhd
What a time to be alive!
Marques said the words but I heard your voice
Ayyyyyyyyyy
Marques reminded me of you when he said that! I scrolled down to the comments and found you :)
Dear scholars hold onto your papers
Now squeeeeze those papers!
This terrifies me more than I care to admit. Not the fact it can make thumbnails, just the amount of articifial intelligence it takes to create these images and how far we've come in the past few years. In ten years the media landscape is going to be so different.
@@rahman_ Bots most likely
You probably haven’t even watched the whole video
Still waiting for the Michelin Pilot Sport EV review 😁😁
Honestly it sort of worries me. People already can't be convinced that artists should earn a living wage and this is just going to make it impossible for freelance artists to make a living. Instead of the classic
"You want how much for a commission? I could just do that myself!"
Artists will instead be bombarded with.
"You want how much for a commission? I'll just use an image generator!"
Yes
As a teacher, this would be incredibly useful for presentations and class materials! I spend way too much time browsing and searching for clipart that I desire. DALL·E 2 would be such a great help and time saver!
This will also kill a lot of clipart websites and businesses that charge too much.
Very true
Also it would most likely destroy most artist/ graphic designers/ painters / people who create banners or brands careers and even grassroots minds starting out on platforms say like Fiverr to develop skills and make a quick buck to help continue down the artistic pathway.. hard to compete with a computer who can develop multiple pictures in multiple styles in a fraction of the time and an infinite amount cheaper if such a program were free. so many positives tho don't forget the negitives😆. Pick your poison..
@@cohwyncubb1572 ive wanted to do graphic design… didnt think about it that way :/
You’re career is obsolete. The AI will teach the human meat puppets.
Ahah we're all watching Two Minute Papers. These new AI stuff are gonna help us make concepts faster. But wait for the day when Tiktok is flooded with viral content generated with AI only
In the future, entertainment will be randomly generated.
- VeggieTales
Deep fakes are extremely concerning, because people believe everything they see and hear. We've stepped into a time now where people need to be taught that there's an equal chance that the video or image they saw is not real.
all the gods are out
There is already a ton of videos generated on TikTok. Like the ones where there is a random 30sec scene from movies or the reddit answers with Minecraft parkour in the background
In George Orwell's 1984, New hit songs are written by computer as part of the strategy to distract the common people.
I’ve had access to DALLE 2 for about a week now and I can’t stop using it. I’ve been trying to see if it can help me invent new product ideas.
I wasn't expecting to see Unnecessary Inventions here lol, threw me off lol
how did you get access to it?
Figures they give access to youtubers and not anyone else 'less than'
ive signed up on 3 different emails and still havent got an invite. I did the same with gpt 3. open ai fucking sucks
@@EddieBurke I got gpt-3 pretty fast (2 months) maybe you’re doing something wrong?
imagine explaining this to someone from the middle ages
They'd probably hang you for witchcraft
calamity
Can't wait for you to make a video inputting 0 3 5 and seeing what comes up. Love you Rudy.
or my parents
witch!!
I love how the the thumbnail you asked for was “a robot hand drawing” at 14:30. That could either be a drawing of a robot’s hand or a robot hand drawing something. Because of the ambiguity, it just decided to do both in the same image
Damn I did not realized it. Amazing!
Well, it did provide 9 other variations for Marques to choose from.
I was thinking the exact thing. This shows signs of creativity.
@@AbhishekMukherjee Not creativity, but logic.
@@jesper164a seeing a human child complete the task "robot hand drawing" the same way dall-e did, you'd call it a creative way of tackling the assignment. How come it's just logic when done by dall-e?
In the future you’ll just have to write the movie script. When you’re done you have the movie. In fact you can watch it as you write it. And change it in real time to try different things. Imagine how Hollywood will be transformed.
You wouldn't even have to write the script
Actually GPT3 would write the script for you, you just tell it the themes you want :)
@The Rest of Us
They could potentially personalize the script. Big companies know everything about us. Feed the information to the Scriptwriter AI and enjoy a movie that plays out differently based on who is watching. Maybe even record the audience to make changes on the spot. Maybe to make horror movies more exciting.
@@unorevers7160 Now we're talking.
@@unorevers7160 horror movies would just get traumatizing
This is both incredible and kinda scary at the same time.
Robots are going to make movies😳
specially those people who don't exist are so creepy.
@@StarDust_1988 maybe those people do exist we just don’t know
Yeah, made me wonder if Marques is even real…🤔
@@npots I thought the same thing lmao
That'd be amazing for college/work presentations. As someone that can't draw, that just looks like a dream
That’s be amazing for college:work presentations as someone that can’t draw, that just looks like a dream
Hello my friends, I hope you can help me. I beg you because of my harsh circumstances and the death of my husband. My daughter and I live in miserable conditions, but my problem is that you are engaged and her marriage is close, I swear to you, I do not want to shame her in front of her husband's family
Eyes
You're thinking way too small.
you'll never get access to it anyway, only select people like influencers get the privilege which is complete nonsense
Marques saying "What a time to be alive" at the end of the video clearly reminds of Two minutes papers channel :D
Hold onto your papers…
I'll see you guys, next time!
thought the same haha
@@MrDevilex94 now squeeeeeze those papers!
Dear fellow scholars
i remember watching this being so mind blown that so could make art. fast forward to today and i can’t escape ai art
Jawbreaker I just watched your video on AI art😂
As impressive as it is scary. Though in fairness the ability to type something in and have an image appear in front of you probably would scare artists in 90s too. But Google Images is already here and leveraged _by_ artists to create new ideas. Maybe Dall-e would would be used similarly. Artists could start projects by generating ideas with Dall-e, then improve them manually. For a lot of people the dall-e result will be enough, but others will want further customizations.
Google images is just searching and finding, it shouldn't scare artists in the 90s, it should scare librarians in the 90s.
Now Google didn't exactly replace Librarians. But in this current age, if I wanted to know and search more about something like "Photon Beams" it would be laughably stupid of me to go to a Librarian before doing a quick google search. I would only go to a librarian once I've hit a wall and really need some help maybe accessing really old books or something like that.
And extrapolating from that analogy, in the future, I would expect that if I wanted a new painting of a samurai fighting a giant lizard, it would be extremely stupid of me to go to an artist before doing a quick AI image generation. I would only go to an artist once I've hit a wall and really need some help fine-tuning the generated image or adding crazy level of detail.
Artist will be not needed... it's not a help for artist- it is artist replacement.
based on Elon Musk, "in future services will be very very cheap..."
(It's weird to Imagine such a world)
@@adamraduso4502 it's more for digital artists. physical ones may last a little longer.
It just pushes artist to be more creative than the AI/algorithm. The images will all start to look kinda similar/style and ppl will either be able to tell consciously or subconsciously as it oversaturates the market (devaluing most artist in the process, yay) while creating a new market for ppl talented enough to either make more original looking stuff, or like you said the ones who can add to it.
I don't like it either way, it just seems like a new way to run old game (automation & capitalism finding new ways to monetize creative space).. Creativity and passion don't come from those things, I'll give you emotion but when automation & cap is involved it's usually negative ones (unless the work itself has a higher meaning/purpose and the automation is helping which is far and few, or the artist is using payment to support themselves or others they care about in some way).
"what a time to be alive" indeed! A video about Ai imaging wouldn't be complete without Two Minute Papers signature catchphrase ❤️
I thought so too
I'm squeezing my paper!
I'm glad I'm young enough that I'll probably get to see this way more developed. Imagine where AI would be in 20 years.
I caught that too, wasn't sure if it was a direct reference or just the visceral reaction to such an amazing research paper and product.
@@dbrophy tighter than ever before!
Artist: "Well, at least I have a job that ain't replaceable by any machine"
AI: "Hold my bit"
Me planning to buy an ipad for art: "oh no.... i'll be useless"
AI: "Let me draw you this beer to hold"
Artists: take days or weeks to finish a piece, ask 1000 dollars, then say "don't repost, don't edit, don't use, don't even look at it" when it's finished
Also artists in a near future: NOOOOO WHY ARE THEY REPLACING ME ??
@@LuiZ-jy1pi it seems pretty clear that you have no idea how artists do things
As someone who can’t draw to save my life but vividly visualizes things (ideas, places, things) in my head and has a knack for detail and descriptions, it would mean everything to me to finally be able to bring those to life in picture, not just words!!! I’ve longed to be able to do that, to describe what I see in my head and watch it take shape in a picture/painting! I know this most likely won’t be available to the public for a long while but knowing there’s something out there like this is thrilling!
Humans: “A picture is worth a thousand words.”
DALL•E: “I can make that in less than 20.”
doesn't really apply tho, because you still need a thousand words to describe that tiger in atlantis photos.
@@elm4nsuri you mean code
@@masol3726 Not code, words, as you would be explaining it to a human. Plus there is so much going on in the photo, that you could easily spend two pages to describe absolutely everything including colors and layout of every single thing featured in the pic.
@@elm4nsuri ...😐
@@josiahjray whats wrong
As a computer scientist, I feel really glad DALL-E is covered by mainstream youtuber like you. It proves Marques Brownlee take a creative process on creating new content. Not just reviewing latest gadget and whatnot.
This is amazing. Terrifying. But amazing tech.
Terrifying?
Can you read my name...
Read my name...
As with any new technology, there are always potential risks and negative side effects. For instance, these technologies could be used to create fake images that could be used to spread misinformation. I am glad Open AI is rolling this out slowly and keeping many prompt capabilities under lockdown or filtered. CLIP will also help in using context to filter.
In future AI will read this comment and also will ask you why did you write that?
I can see Dall-E supporting artist productivity for a long time to come instead of flat out replacing them. The ideas you can play with before you create a finalised product is insane. You can also tackle projects now that would have been way too expensive to do before and often were simply unfeasible.
Uh basically you're saying artists will just pretend that Whatever Dall-E produces was their own work. That's bullshit, and still means the artists were replaced. The artists are just being dishonest if they do this, but they are still uneccessary, it is just a matter of time before their employer figures out what they are doing, and then they will actually be replaced.
This tech is actually insane. I would use it to brain storm my own ideas: A Burrito made in the style of JoJo's Bizarre Adventure.
Populate the image.
Then go make the burrito.
Yeah
Of course, it would be Jojo.
As an illustrator and tech enthusiast, this terrifies and amazes me at the same time.
same
Our end is near my friend.
@@damon95 they took rrr jobss!
As an animator and designer, i beginning to understand how truckers feel.
Lol
Haha
Ahaahahhahahaahahaha
🤣🤣🤣
For real though..
I have been playing around with AI art a bit and I think the technology will replace a lot of things. I don't think it will replace the culture of art though. I watch a lot of artists on UA-cam and the reason I like it is because of the skill. When I watch someone draw a picture perfect image it makes me excited, and that is what gives it value. I don't look at my printer in the same way.
Wait until they make another ai that generates a speed paint of the art made by the first ai. And then they make another ai which makes a voice over for the speedpaint.
@@somethingclever4297 Humans will always have a bias towards other humans. The printer didn't kill the price of art like most thought it would.
@@SmudgeOfficialUK I was making a joke.
@@somethingclever4297 ahh with ya. Sorry mate. Hard to tell on youtube these days lol.
@@SmudgeOfficialUK I did try to hint at it with the "make another ai to do a voice over."
DALL-E 2 is an inflection point in human history in the creation of culture and art, and there is more to come in the following years thanks to the Deep Learning revolution. Awesome to see everyday more people talking about this!
Cool
Frr
Don't you mean "evolution" instead of revolution?
Closer and closer to "the singularity" every day. *actually* insane. The next 10 years are going to be very very very interesting.
Que bueno verte por acá Carlos! Soy fan de tu canal :) Saludos!
thank god someone is actually talking about this who has a platform. i felt like a crazy person the more gpt and dalle developed to be absolutely op and no one talking about it like it wasnt crazy human achievment
Me too, bro
Totally… since the recent updates (2-3 month) and me becoming aware of the improvements across the board… I was running around and telling everybody that general AI is just around the corner. I got mixed reactions from disbelieve to awe when demonstrating stuff. It is just shy of being so convincing that you forget you're prompting a machine… and my open AI bill skyrocketed as I just couldn't stop using the thing.
As someone who is just getting into AI engineering holy shit. I have a long way to go...... It's discouraging but also fascinating to see. I wanna get to this level.
Check out the channel Coldfusion. He made a video on DALL•E 1 over a year ago and made a new one a few weeks ago about this version.
Lots of interesting history of tech stories over there as well as introductions into new technologies.
@@maxziebell4013 same, i had to stop using the gpt3 playground because i was spending too much (i may or may not have been using it to finish assignments and essays)
While this might look like it's gonna replace the jobs of artists, I feel like it's an *extremely* helpful tool for artists. Being able to get a reference image of something that doesn't exist, in under 10 seconds. Amazing
In its current iteration it's more of a quick sketching tool unless you ask for something pragmatic with a fair amount of training data but that's already mighty impressive and useful for some applications like storyboarding and who knows where the technology will be a few papers down the line :D
I use the AI app "DREAM" just to get ideas that I can then expand off of. AI is actually very helpful for looking outside the box, or giving a base.
It will be helpful for some, it will replace a lot. And that’s only immediately. The better it becomes the more it will replace. I’m 15 lol. I wanted to be an artist and now I don’t really have a future.
I have been playing around with AI art a bit and I think the technology will replace a lot of things. I don't think it will replace the culture of art though. I watch a lot of artists on UA-cam and the reason I like it is because of the skill. When I watch someone draw a picture perfect image it makes me excited, and that is what gives it value. I don't look at my printer in the same way. Often is actually a process of connecting with other humans. Maybe one day robots will get there.
once these tools tap into other systems like gaming engines such as unreal engine and similar, video games might be popping out super super fast, not to mention movies once this technology can do the same for video.
As a graphic designer I was thinking the same thing before you brought it up... I feel like my job will be obsolete by automation, or something like this. At the very least I would like this tool so that I can get inspiration, or use an image from it and clean it up to be used
I don't think a graphic designers job will be obsolete - maybe you will need to adapt your workflow a bit and utilize new tools, but in the end I don't think that an AI could fully replace the creative head of a graphic designer.
Imagine 10-15 years in the future desperately wanting a fanfiction film of a franchise you love and ask your computer in detail to make it in the length and with the story you want. That's incredible! You could create your own media specific to you without relying on studios to deliver.
It’s like the Star Trek holo deck
This reminds me to a series where in this series the tv programs are only computer generated. And a shit series with a baby is the most watched one because it's generated to be addictive. Maybe Rick and morty? I don't know anymore
sounds more boring than amazing
John Wick in Marvel universe maybe ☺️
That will be amazing to visualize fantasy stories, turn any book into a movie according to your preferences, flesh out your roleplaying stories for games like D&D and so much more.
The pace of progress in AI is astonishing and I'm both very excited and very terrified of the future possiblities.
Therefore I'm glad OpenAI is not as open as the name suggests.
That Burger “Kringe” 😂 12:22
"What a time to be alive" is often said by Dr. Károly Zsolnai-Fehér on his UA-cam channel Two Minute Papers. I highly recommend checking it out if computer graphics and machine learning are interesting to you. Each video is a summary of an academic paper in those areas. It's been really cool to watch the progress of these things over the last few years. The jump from Dall-E 1 to 2 was way bigger than I expected.
He has covered a few NeRF (Neural Radiance Fields) papers which it seems finally made it to a consumer product with some of the future Maps features Google showed off at I/O. Immersive View and the restaurant fly-through seem to be the same or a similar approach applied to their billions of street view and indoor imagery. Definitely looking forward to checking it out myself.
I was about to say the same thing :D I think that must be a reference to him
Hold onto your papers!
yes, i've seen that video! great channel.
Pretty sure that was an intentional reference to Two Minute Papers.
You are too kind - thank you so much for the shoutout!
Ive been making heavy metal album covers with ai art, it’s pretty gnarly
As someone who makes video thumbnails for a job, this is one of the scariest things I've seen on my homepage
Rip Jay
Be prepared to have your comment bombarded with a bunch of people explaining to you why you're awful for trying to make living off something that isn't a way they personally approve of and how happy they'll be to see you lose out.
People in the comments of tech videos are very predictable.
AI will take all jobs sooner or later. If you can make it draw any picture, make music, direct any movie, then you can also make it do anything else as these creative things are the peak of human capability.
@@bezosphere I'll see you ina few hours when this post has a bunch of replies doing exactly what I said lol until then peace ✌️
@@strayiggytv uh what? Why would people not approve of his job?
I remember when people used to say that computers might be able to do repetitive work but never be able to take on art like painting, writing, etc. But I always knew it was just a matter of time until the AI got better.
Can't wait for the day we can all input our health issues into an AI like this, and the results will spit out working solutions. Even to currently incurable diseases.
How much of the images generated can be considered original content, and how much is just a stylized copy of what was in the training data?
@@mickaelzehren8249 In a sense every human artist is also only producing art that is based on their own training data, a lifetime of training on references, impressions, art concepts, and styles. In theory an AI could build on millenia of human experience, much more than any single human could ever be aware of.
"I remember when people used to say that computers might be able to do repetitive work but never be able to take on art like painting, writing" - now, here's the question. Is a neural network simulated in a computer still a computer per se. Because in my opinion it's not, so the statement stays correct.
@@mjanek20 Yes it is. The neural net is running on a computer / server hence the statement is false.
@@mjanek20 I think a neutral network has to be seen like a computer. In the sense that a neural network behaves like a mathematical equation, it is completely different to how our brain works. For example, our brain does not work in a synchronous and discrete manner like the "neurons" of dall-e 2
14:07 After effects used to have a brainstorm button that would basically randomize parameters on an effect, and it would give some interesting results. They removed it, but I thought it was kinda need for generating new ideas.
How long to wait until access is granted?
@@melamoloi555 still haven’t got it.. all browerserv language are needing correction this post was1 year ago
Art itself is something more than just the result, I realised it when I entered the pompidou centre in Paris. There I saw a ton of modern and contemporary art pieces and I was shocked: the amount of creativity people can reach is incredible.
There was this piece of a painted water fall with a shower on it. Like a real shower. Builted into the painting. The idea of putting a shower on a painting is what makes it genuine and artistic, it's not the final result, it's not something you can put on a wall, it's artistic. It's artistic because it's what the artist thought of doing that makes it artistic.
You see you coukd ask sonething like this to dall-e "make a contemporary art of a waterfall" but it will never show this result, it will make some random styled normal waterfall. Sure you could ask "put a shower on a painting" but if you don't ask for it you'll never find it.
Fact is that humans will always be the motor in society, watching stuff and having smart and clever ideas from what they see, we are able to make this xonnections alone without someone telling us and that's why we will always be the artists. Ai copies, replicates, imitate, it doesn't "invent" anything it just smash stuff togheter without thinking. It's us, with our imagination, with our inspiration that transform that image into something real.
We will always be the motor. If you think this is "a computer making art" imho you don't know what art is. Art is much more than connecting two dots togheter.
such an underrated comment
I'm pretty sure AI could come up with the idea of putting a real shower inside a painting.
Imagine a tool that can make storyboards, in wich you detail things like characters, angle, height of the camera , movement, action and the AI generates sketchs of the scene
oooo now that's an idea !
For now it' s called a 3D software
wow ur idea is revolutionary, u could pretty much write a whole manga, anime, story, movie, etc.
Maybe we were living in such a simulation. Imagine what a planet sized computer could do
@@ulysse6916 but can it give you variation of sketches in mere minutes even seconds?
This is actually monumental. I’m sure soon you can ask AI to write a whole script based off a concept and have it not only deliver a great story, but animate an infinite number of episodes on the fly.
scary though, so many jobs being killed. I don't think AI will ever kill human creativity though.
@@williamd7062 yeah I’m definitely not convinced that it’s a good thing
@@williamd7062 it will
This is the plot of a Roald Dahl story. I think about often when I see discussions about AI. In fact I would say it's what made me want to study philosophy when I went to university. I highly recommend it, it's in the short story collection called "The Umbrella Man and Other Stories," and actually you can read the whole story in the Amazon preview if you want.
@@Osirus116 we’re witnessing the death of human art. The opposite of the renaissance
Stuff like this is far more interesting than a new phone review which basically boils down to "better camera, slightly better performance"
I think those videos end up paying for stuff like this to some extent. A review of an iPhone is less niche than a video about advanced visual AI.
Already has 1M views in 1d though. Not that niche. Reminded me of the camera robot, just cool new interesting tech
Hey, Friend, although this is a very cool instrument indeed, a lot of people going to loose their job because of this and AI don't really create any new job that they can shift to.
And this applies not only to their job, but also to YOUR job, my friend, who is reading this now. (Including bloggers)
Soon we'll get AGI good enough to take all jobs. in our consumer society AGI can't be properly regulated and will be most likely monopolized, this will lead to world dictatorship of few/one man. To assure that this won't happen we, regular people, need to unite and build Creative society. Check out this project.
AI monopolization crisis is one of crises discussed on previous conference "Global Crisis. This already affects everyone"
And on may 7, international forum organized by volunteers from 180 countries "Global Crisis. We are People. We Want to Live" took place with 100 languages of simultaneous interpretation. People from all over the world voiced the truth about how we can solve ever-increasing crises of consumer society and build Creative Society🌍🌏🌎 I'm very happy to invite you from the bottom of my heart💗
We still do really need to improve technology, and it'll be improved anyway. So the point is that we need to improve our society accordingly so technology improvement in it will only lead to everyone's good.
this is so insane, its going to change the world in a lot of aspects especially with art.
Other than art, how?
@@wavygoods877 design and engineering soon.
@@RB-mm7ce i think this proves that bring an artist is going to become useless?
Negatively
@@PeixeDeEstoque ofcourse
I've been playing with a bunch of these tools for a few weeks (Haven't gotten DALL-E access yet) and hoooo boy there's so much to unpack.
which tools are available for public access
Can you recommend any
please?
?
Put me in the loop
Imagine a future where "on-demand" entertainment is created on demand the moment you request it.
Jesus, imagine the affects of that on our already damaging dopamine addiction.
It is already happening!
14:59 missed opportunity to cite TwoMinutePapers :D
Content and context 👇🏻
10:40 Limits
11:35 Quirks
12:15 Letter error has been mostly fixed now
12:35 image transformation
13:17 taking jobs fast - brainstorming
15:00 Two minute papers
imagine asking for a movie in a specific genre and it produce one for you exactly the one you want
Shhh Netflix will make you disappear. But no, there's already something similar to this I believe Amazon was building back in their scratched writing development program. They were asking for ideas and ended up stealing a bunch. The logical thing for them to do is to build something to do this. That's why most movies suck now. Think about a system similar to scattergories, but random stuff. High grossing movie + new location. Titanic + in space = Passengers.
There are already AI that can write scripts for years now and they're not even top secret fancy stuff but something anyone can pickup and train. They're usually hilarious because grammar and human emotions are hard to do but I assume the more advanced ones are really close.
The language understanding it’s probably the most impressive part. This is what holds us back from the future with a Star Trek AI we could talk to, looks like it’s closer than I thought
Language processing has gotten so good now that you can literally talk to google voice assistance and have a decent conversation. 5 years ago, it was pretty much impossible.
10:57
r34 artists: DANG IT!
This is so dope, and kinda scary tbh. AI is becoming so advanced
What lots of people fail to realize is that if this is where the technology is now, the exponential advancement of technology means that’s what is coming next is going to be unbelievable, and soon
I just can imagine some AI like DALL-E 2 being able to create a song from scratch, just giving some descriptions of the song, rhythm, style and even a known singer voice. That sounds pretty tangible and would be absolutely impressive too 😳🤯
"Photoshop" for voice and sound has already been made several years ago. You can mimicing someone's voice by giving it a small sample and be able to create a TTS based on his voice. It's both amazing and scary.
I love how “Person of Interest”, in 2008, guessed what computers could do in the future. As that show has an AI that created a whole new person, picture and all.
In the future, this is how movies will be made. The talent would be the creative team actually describe the movie so detailed enough to manipulate and add variables to the AI.
Nah, AI will just work out what the viewer wants and give it to them, no humans required.
He said the moon wasn’t perfect. Reminded me of iRobot when Will Smith asks the robot if a robot can conduct a symphony and the robot replies, “can you?”
the AI keep the golden ratio, colour pallet and combinations in check too. Impressive
I've been obsessed with following new DALL-E 2 prompts recently. The sheer potential of this kind of illustrative AI is magical! And what's funny is that media synthesis is such a young field, barely a decade old and it's seen leaps and bounds in everything from text generation to voice synthesis to image generation. I remember when face generation in 2014 was grainy and out of focus and we've already cracked photoreal faces 3-4 years later.
Heck, we're at 1080p pictures right now? What about 4k next year? The diffusion model it uses has been used for short gifs already, so we're only a matter of time from videos. Movies. TV shows. Games! Entire VR worlds!
We're on the cusp of a creative revolution and an utter from-the-ground-up democratization of art!
How does this democratize art? Sure, it makes it for you, but doesn't teach u how to make it. It doesn't teach u perspective, or color theory, or composition, or illustration, etc. It just gives u nearly-finished images.
im confused how this democratizes art? It's just going to put artists out of jobs when corporations realize they can pay a yearly subscription for an AI instead of paying human beings salaries and keeping up with their well being
There is already a fair amount of a.i generated music on Spotify apparently and that crazy a.i that can write news articles etc could write novels I think. When we can't really tell anymore what's what I suspect human creativity will go down rather than up as we lose context for reality and general consciousness is corrupted. It will be interesting to see what happens. Feels like it could go any number of insane unforeseen ways when all simulation we receive from digital media is not variable.
@@sor7en07 It lets more people create by lowering the barrier to entry to more people. Not like modern painters aren't less of an artist because they don't mix their own paints, or digital artists for having a undo button. Difficulty should never preclude creation.
And there's still a good amount of direction and prompt engineering that needs to be done for AI tools like these. Indeed, professional artists who know what to ask for make some of the best prompts I've seen for Dall-E.
@@FedThePoopy OpenAI isn't the only one doing art tools with AI. VQGAN+CLIP, StyleGAN, DiscoDiffusion, and Midjourney are some of the big ones that are open source and commonly iterated on. Just recently open source language models like GPT-NeoX surpassed OpenAI's old GPT-2.
Heck, the whole modern media synthesis revolution in the last decade with deepfakes first hit the news with some guy making deepfake porn himself. While open source tools like these are about a year or so behind the quality of big corporate ones, it shows that this isn't something constrained by corporations alone.
If you compare this to any previous AI models like CLIP or VQGAN, even DALLE-1, the progress is mind-blowing.
Great now everyone can make NFT-s with this XD
fr
Oh wow its NikTek xD
love your memes
Now no one will buy them! :D
Hey I had the same idea 😄
But are those pictures really Copyright free?
people have been saying for sometime, that robots with A.I will handle all the hard manual labour and humans can just focus on more creative stuff in the future....but it seems A.I does a better job of creative stuff than humans, it won't be long until we see art, clips, short videos and eventually full movies made by A.I...
Really? It just uses the art that humans came up with, but not invent new styles or techniques
@@therpope what other techniques does even exist that humans have to done? Like we did art since stone age. It's like complaining "but oh, the ai will always make the wheel round like a circle"
@@therpope well yes, but it can do it better and faster....
you can probably create an entire anime or a comics for that matter using this in a day or so....
Corridor crew tried making a social media influencer using only AI. It's not perfect but the results are amazing still.
None of this is CREATIVE. It’s just adding already present images. It can’t “create” new images.
7:10 so manny interesting details
Within 10 years we will have software like "Make a comedy movie with Tom Cruise dressed as a t-rex"
This is both fascinating and terrifying at the same time!
From a tech-geek perspective, it's incredible how this even exists, but from a human ethical standpoint, it causes enormous concerns.
I can already imagine DALL-E 200 synthetizing/manufacturing/printing objects or useful machines based on the user's description of requirements, the impact it would have on manufacturing :- O
I think it won't take 200 versions
My Friend, although this is a very cool instrument indeed, a lot of people going to loose their job because of this and AI don't really create any new job that they can shift to.
And this applies not only to their job, but also to YOUR job, my friend, who is reading this now. (Including bloggers)
Soon we'll get AGI good enough to take all jobs. in our consumer society AGI can't be properly regulated and will be most likely monopolized, this will lead to world dictatorship of few/one man. To assure that this won't happen we, regular people, need to unite and build Creative society. Check out this project.
AI monopolization crisis is one of crises discussed on previous conference "Global Crisis. This already affects everyone"
And on may 7, international forum organized by volunteers from 180 countries "Global Crisis. We are People. We Want to Live" took place with 100 languages of simultaneous interpretation. People from all over the world voiced the truth about how we can solve ever-increasing crises of consumer society and build Creative Society🌍🌏🌎 I'm very happy to invite you from the bottom of my heart💗
We still do really need to improve technology, and it'll be improved anyway. So the point is that we need to improve our society accordingly so technology improvement in it will only lead to everyone's good.
Finally I can describe my weird "Dreams" that I get while I'm asleep, to someone!!
This seems like a perfect tool to generate ideas so artists can get more creative. But I also believe that in the future you would probably be able to tell an AI which parts of the image you don't like and how you want them altered and iterate very fast this way. At the moment you can only really get to one iteration, when with a graphical designer they can make modifictations for you. It will probably eventually make some people lose their jobs, but new jobs will come in its place.
Edit: it already is possible to tell an AI which parts of an image tou don't like and iterate with it.
This is the most “mind blown “ tech I’ve ever seen. Period. I appreciate it even more as an artist. Just incredible.
Then brace yourself for Gato from Deep Mind, it's the closest thing we have for a general AI and it's already stealing the show from both DALL-E and GPT.
@@Shadi_Wajed can't wait to see that
When you brought up the Mona Lisa goat results, I said "oh my god" out loud. This is incredible. I also totally understand why a tool like this is a very long way away from general public access.
Well as long as it remains limited to illustration, I see no reason why they should further regulate this program, given that the majority of people believe it is essentially fake. On the other hand, I concur with his assessment of photorealism. There should be some regulation to ensure that it does not cause further harm, as the majority of people will have difficulty determining whether it is real or fake. Again, regulation should apply to photo realistic generation only, not illustration.
@@mr_sb the problem is the amount of computational power it requires I think. They have to run this on a server, it can't run on your personal computer yet. If you suddenly allow billions of people access to it at once things wouldn't be pretty
@@mr_sb They're only in control so long as this kind of software remains in the realm of corporations. Following this field for a good while has told me that open source and accessible tooling follows pretty fast. Just recently, EleutherAI matched OpenAi's GPT-2 with an open source equivalent. Hell, DeepFakes first hit the news when some guy in his basement was churning out porn of celebrities by himself.
@@alexdoan273 wouldn't mind having it on my PC but it took 5 minutes to calculate.
@@HAWXLEADER I think it takes massive amount of storage too, other than time. Because of the amount of training data. Just a blind guess
while this is certainly amazing and mind boggling, a part of me has a sense of doom about AI like this. like, we will eventually get to a place where we no longer can tell what is real or AI anymore, and it will be so pervasive we will eventually stop wondering/caring. in essence, a reality fatigue. i’m learning to appreciate our time a little bit more knowing how much we make with our own hands, artistically and creatively, and as creators this may be our Golden Age. what will our descendants reality be like? kind of scary/sad to think about.
There might be pushback if it goes further enough, perhaps like a legal push toward some sort of identifying system, or maybe a mass unplugging and embracing of the physical world where you know things weren't AI influenced.
@@h..h because humans don't relate to machines as much.
@@h..h Because governments can create fake photos and accuse other countries of reckless or criminal acts to justify certain responses, or use it in other ways. (Eg fake photos of Presidents doing illegal things)
@@h..h For example, deepfake content can have serious implications on truth-telling.
Hey, Friend, this is a very cool instrument indeed, but a lot of people going to loose their job because of this and AI don't really create any new job that they can shift to.
And this applies not only to their job, but also to YOUR job, my friend, who is reading this now. (Including bloggers)
Soon we'll get AGI good enough to take all jobs. in our consumer society AGI can't be properly regulated and will be most likely monopolized, this will lead to world dictatorship of few/one man. To assure that this won't happen we, regular people, need to unite and build Creative society. Check out this project.
AI monopolization crisis is one of crises discussed on previous conference "Global Crisis. This already affects everyone"
And on may 7, international forum organized by volunteers from 180 countries "Global Crisis. We are People. We Want to Live" took place with 100 languages of simultaneous interpretation. People from all over the world voiced the truth about how we can solve ever-increasing crises of consumer society and build Creative Society🌍🌏🌎 I'm very happy to invite you from the bottom of my heart💗
It's amazing that this video was only uploaded a little over a year ago, and yet it's completely out of date. The way Marques talks about ai as well feels like it's from a different era.
Totally, came back here after watching a video about AI influencers...
This is what I expect from a tech channel and I love it!
This is so cool!
You have an audience you can try it.
Future is looking ugly bring me back to the 90s
Hey, Friend, this is a very cool instrument indeed, but a lot of people going to loose their job because of this and AI don't really create any new job that they can shift to.
And this applies not only to their job, but also to YOUR job, my friend, who is reading this now. (Including bloggers)
Soon we'll get AGI good enough to take all jobs. in our consumer society AGI can't be properly regulated and will be most likely monopolized, this will lead to world dictatorship of few/one man. To assure that this won't happen we, regular people, need to unite and build Creative society. Check out this project.
AI monopolization crisis is one of crises discussed on previous conference "Global Crisis. This already affects everyone"
And on may 7, international forum organized by volunteers from 180 countries "Global Crisis. We are People. We Want to Live" took place with 100 languages of simultaneous interpretation. People from all over the world voiced the truth about how we can solve ever-increasing crises of consumer society and build Creative Society🌍🌏🌎 I'm very happy to invite you from the bottom of my heart💗
We still do really need to improve technology, and it'll be improved anyway. So the point is that we need to improve our society accordingly so technology improvement in it will only lead to everyone's good.
There is an AI that's talented at finding deepfakes due to photo editing artifacts and poor lighting. I wonder if the anti-deepfake AI could teach Dall-E where its images keep failing, thus making Dall-E a better artist.
We're doomed when AIs start teaching each other stuff. That's... almost like how humans learn.
That's exactly how these AIs learn. In Generative Adversarial Networks a network generates an image and another tries to find flaws in it. The one that generates learns by improving itself to the point that it can fool the other one.
Hold up let's take a step back, there's an anti-deepfake AI?? God bless whoever created that.
In other words: Yes, this will totally take away people's job. Though, it might take a couple of years more. As you said, it will get more advance and spit out more sharper text/images. High quality photo's and video's etc etc... For now, i don't think we immediately need to be "worried." As you said, it is a very good concept for things like when you are brainstorming and having images in about 10 sec. to see what could be done. But of course, you would want a specific high quality image in the end for the project. Which it can't do (yet.) But i guess they're very close. The thing is, it isn't available for most of the public. So if it will be mostly for very high level professionals like big corporates, well, let's say we still have a while to go.
On the other hand, there are still people who just like to design their own stuff. People like myself who actually do like to design. Of course, when it comes to pro's for their jobs. Time is money. If you can make images in les then a couple of seconds. They won't hesitate to get this stuff and let them replace you for most of the part. They might still have some graphic designers hired to kind of control things i guess.
But yeah, what a time to be alive. (In some cases.) I bet that, when they first found out the lamp and the telephone it was also a great time to be alive. I remember having one of the first cellphones and then the smartphone. Now that we can see kind of its peak design when it comes to smartphones. Well, the question becomes always: What's next? Like for tv's... We have the Micro LED from Samsung. "The Wall." Which looks like a nice (but very expensive) concept!
I will have access to Dalle-2 next week, I am planning on doing some weird camera designs that never existed
that would be awesome
I want to see the results
Cool idea!
Please ask it to draw itself ... I'm haunted by what the result would be ...
Ask like superheroes in a bowl of supervillions
This is scary good. Maybe in few decades, general public can use it like Google Image search, selecting text in Image, Image to Text & Scan Text.
Based on how it handled the camera, and wall of PCs, it seems like this might be better suited for organics, or heavily stylized art. Either way, I think this is really cool, and would love to play around with it.
I mean, the wall of PC's were way better than when it tried to turn the iPhone to older technology. In the end it's AI and not a reasoning entity that can actually understand the concepts it gets fed. For instance, the first blown up dog image looked really off to me on the right side. Pretty impressive all in all though.
Not really, it pulls off photorealistic images without a problem, most of the time, and when it doesn't, the issue isn't the presentation, but sometimes it gets a bit confused and starts doing whatever it wants to.
13:35 - The jobs are kind basic the same but I see them both working together instead of competing. For example, instead of committing and putting effort for your graphic designer, you could go with dalle and show to your client multiple possibilities, then you come back with the chosen one and your graphic designer will create something on top of that. What I see is more of a shift of creativity going closer to the customer so they can share feedback and expectations early.
Imagine feeding this AI a whole novel and seeing how it illustrates the whole book. Also, a consumer product for generating very specific pictures for powerpoints/flyers would be awesome.
I imagine giving this a page or two of Tolkien describing a room. Would be so fascinating to see what it would spit out.
@@thewaterboy2013 the whole 'living in a simulation' thing is starting to take form & make sense. Infinite possibilities...
I’d seen the first version before, but this is next level. I’d better find a new job
It's really sad the amount of people who will probably lose their jobs because of this. It reminds me of when the cotton loom was invented. It stole so many jobs, that people rioted and destroyed the machines. I guess that the advancement of technology is inevitable, though
@@watermelon4558 As you say, it's kinda inevitable at this point, because anyone who's smart enough can do it. It's open to anyone. And so we should debate and think about the future that's gonna come. We should brainstorm ideas and collect them for situations like singularity, where AI will rapidly start to accelerate intelligence to the point where we are like ants to them. What then?
@@watermelon4558 Nothing like the fury of those who believe in the Divine Right of Stagnation.
@@AndSendMe I never meant to say that technology should be stopped. I meant to say that despite destroying jobs, technology also creates new jobs and technology should be protected
@@watermelon4558 I was referring to the machine-destroyers, but when we think that loss of jobs to advancing technology is sad, we share in the idea that life "ought" to be easy and stable, and shouldn't require us to keep our eyes open and be ready for change.
An easy example to understand what a general AI (often spelled AGI for artificial general intelligence) would mean is that it can start without what a painting is or how one can draw in the first place, but you could throw fully 1000 scanned art books (or PDF files) at it and after it has learned about art and drawings from those books, it could do DALL-E 2 stuff without any extra target specific programming.
Now imagine such AI system watching every single tutorial video in UA-cam and understanding it all...
2 minute papers just posted a video of exactly this. Google's Deepmind AI
@@jenkem4464 As far as I know, DeepMind Gate cannot generate images similar to DALL-E 2. It can do any of the 600 tasks it has been trained but I've yet to see any evidence that you could give it collection of art books and then ask it to do something similar to DALL-E 2 or even DALL-E mini.
@@MikkoRantalainen Yet.
I feel horrible for people who have spent decades honing their craft at digital art. It may not be a perfect replacement, but its definitely going to make a difficult and competitive field even more challenging.
Nothing stopping them from using the technology as well. A text prompt only allows for so much granularity in how complex an image can be.
@@whwhwhhwhhhwhdldkjdsnsjsks6544 that's true, but finding work is the hardest part in any creative industry. What happens when your boss can just smash his sausage fingers on the keyboard and instantly get a competent graphic? Why pay someone to do it? I know you pay for expertise and a good eye, but a lot of business owners would rather save the $50k+ per year
@@NormanTiner yeah, that is the big issue; in a society where everyone is expected to work then even if money is being made not everyone will be paid. This is the thing that most concerns me; as far as we know we can’t really change to anything other than capitalism without crashing our economies, despite that if AI take jobs there is no logical reason that the people displaced by it shouldn’t be able to be paid for it.
Have to say that is really amazing. So someone with little to no drawing skills could flesh out a new product design ideas by just describing what it will do? Cool!
Yea, ad and marketing companies won't need designers or an art department anymore, maybe just one person with art and design background to approve and select DALL-E's creations.
You can't say "What a time to be alive" without saying "Now, squeeze that paper!"
It's fascinating and scary all at the same time, on the one hand I can see this being an amazing tool to use, on the other hand, speaking as a designer, I can see this tech taking people's jobs further down the road.
maybe in 50 or 60 years people will only be able to make computers
Isn't a designer job to compose and improve it til the requirements are met. Make things work together and not create random images all the time.
@@hombacom These are not random images, they are composed, and they can be adjusted for solutions design faster then any designer with a copy of photoshop?
@@sparkplug964 You are still limited to the current algorithm. Is it hard to understand that designers already do and will in future combine AI and editing tools to make a better job?
When you asked about a tiger and Atlantis I was awed particularly by the first image you showed. It's crazy good. I would imagine a real artist creating such image in a span of weeks. It really blew my mind of the capabilities of such powerful AI.
It makes me wonder if in the future should I employ a person to prepare me an artwork I wish for, or whether just a work of AI will suffice. If I were an artist I would definitely worry about my future prospects of work, especially if I happened to work out my distinct art style and AI could just learn it and "steal" it without me even getting noticed, let alone benefiting from that. Crazy world we are living in.
With some refinement of the technology, just imagine this asset for animating D&D campaigns or other fantasy driven naratives. So much potential!
04:19 It's a content creator and it can troll us too? Now I'm *truly* scared.
Amazing stuff, sad that we can't all test it now lol. Definitely seems better for brainstorming rather than finished products, which would helps artists ideate or help clients explain what they want. The images it makes looks more like a first iteration of an idea, pros would usually make more iteration and improvements of those with specific feedback until a final product is achieved.
In the case of art though, I think it could potentially replace the generic commercial art like the stuff printed on random cheap items, gift items, t-shirts, mugs, souvenirs etc. It's interesting to wonder about a world like that, reminds me of the anime series Carol & Tuesday that explores a future where AI generates all the mainstream music with people occasionally singing the generated lyrics, and a couple of girls who try to write & create genuine human music in such a world.
Dall E Mini is available to everyone by now
Incredbile how far we've come in 7 months. And with text to music and video AI the future looks amazing
Tech feels like it’s evolving pretty fast yet pretty slow at the same time doesn’t it?
No it's just going fast
And it's about to kick into another Gear. Shits gonna be crazy
It's moving fast, but it's happening in some places all at once that you don't realize it's happening.
Hey I was having fun 3 days pnp sorry my mental health was tired but I know these since then pls stop I can see my strezlkized not one day job I want speak dealing this then you have love life
In all honesty, I would most likely just use this program to create new wallpapers lol. It's an impressive feat and seems extremely cool, I'm just in need of new wallpapers because I constantly switch them up.
This reminds me of when I was a kid and google earth was new, you'd just play around seeing what you could find
This almost makes me think that it’s some kind of elaborated late April fools joke.
Like I would love to have access to this it looks much more powerful than the normal Dall-e
Wow…🤯. So much I would love to try and those safeguards definitely make sense.
Did the horse have a NASA logo on it?!
I loved the nod to 2 minute papers in the end!
Great video as always, thanks for all the content and information through the years 😄
DALL-E 2 is incredible and I'm very excited to try it one day. Glad that you made a video about that, it deserves the recognition and support from people, the team working on it is spectacularly smart and creative. ✌️❤️
I could watch these generations for hours. I want more
This begs the question: who ultimately owns the pictures generated? You, the person that inspires the creation? Or the AI that generates the image?
The AI can't have rights because it's a machine still. It's nothing like an animal brain physically.
The US patent office has declared that no AI created works can be patented or copyrighted.
AIs themselves cannot be given copyrights, but it's legally unclear whether humans involved in AI works can be given copyrights.
I remember Jukebox (another OpenAI project) launched back in 2018 that could *literally* convert lyrics into actual songs made by your favourite artist. It was insane.
Really? Does it still exist?
13:00 something else that I feel would be actually significant is feeding in a family picture of a lost loved one, and with the help of the AI, creating a new family photo that was otherwise not really possible because well, they aren't around anymore ya know :/. Atleast so that in that fashion, if in life you didn't have the chance to get a frame with all your children or parents, perhaps you may now?
you can already photoshop that
I'm currently doing research in the field of materials informatics and we are using diffusion models like this to generate compounds and chrystal structures. So cool!