Philip has such an abstract idea of a bed, that even DALL-E 2 can't handle it
@BonJoviBeatlesLedZep
2 жыл бұрын
I'm still in doubt that that's his actual bedroom and he comfortably sleeps on that. He lives with his girlfriend right? He must surely be sleeping in an actual sane bed with her most nights.
@YTnamesAreBS
2 жыл бұрын
This comment made me exhale quickly from my nose, which is the highest honor I can bestow on you.
@Blox117
2 жыл бұрын
i tried inputting "honest woman" but dalle 2 told me to ask for something reasonable
@iR0CKt
2 жыл бұрын
@@Blox117 Need to be stylized or something :D
@ChrisD__
2 жыл бұрын
@@iR0CKt That's right, you can only get yourself a furry or anime GF.
@cortster12
2 жыл бұрын
You know the crazy part? We're not reaching the endgame, these tools are still in their infancy. We're like people looking at room sized computers in the 1950s and being amazed. Because it was amazing, the future just ended up being unfathomable.
@vadiks20032
2 жыл бұрын
aren't they written in python imagine the speed if they were written on C languages
@fueledbycoffee583
2 жыл бұрын
@@vadiks20032 written in python calling c++ code. So would be about the same.
@4rumani
Жыл бұрын
this is the end lol, the ai boom is over and we're headed towards a long long ai winter
@cortster12
Жыл бұрын
@@4rumani I will remember this comment when it's completely wrong.
@BlackParade01
7 ай бұрын
@@4rumanioh boy how wrong you were
@TilW
2 жыл бұрын
I am quite impressed with DALL-E 2, but when closely compared to DALL-E Mini and Midjourney, it still falls behind when it comes to one thing: The inablility to generate Boris Johnson in a bath of beans without violating the TOS.
@mkontent
2 жыл бұрын
This.
@mkontent
2 жыл бұрын
I was honestly shocked that Dall-e mini even knows the faces of popular people. Considering how fuzzy the images are, it will still try its best to recreate Boris Johnson, Ryan Gosling, etc. Literally blew my mind. Like, recognizing faces and the names behind them is something a baby human can do. Little AI that knows Ryan Gosling...
@3n3j0t4
2 жыл бұрын
@@mkontent dalle mini is literally the only one that allows faces
@3n3j0t4
2 жыл бұрын
@@mkontent I know you read this you sassy dumbfuck
@explosu
2 жыл бұрын
@@mkontent TBF, Boris Johnson already has a face that looks like it was generated by Dall-E Mini so it probably doesn't struggle as much.
@juliann149
Жыл бұрын
It's crazy watching this video only 9 months later, seeing how much the generators improved already. Would be interesting to re-run the comparison on the current versions, especially Midjourney advanced a lot iirc.
@Tofuey
6 ай бұрын
Even further a year from this comment
@emperorpalpatine1469
2 жыл бұрын
Mate I'm so glad you're still making videos like this, you're probably my favorite chap on KZitem. You taught me how to play counter strike and you got me into technology from a young age, thanks a lot Mr Phillip :)
@llave8662
2 жыл бұрын
DALL-E can't generate words to avoid falsifications, similarly to the reason why it does not allow faces. Great video!
@LazyBuddyBan
2 жыл бұрын
thats explains it. but also won't matter, since likely in 5 years we will get it without restrictions.
@manfail7469
2 жыл бұрын
@@LazyBuddyBan christ, can you imagine how much the world will change when stuff like dall-e 2 go unrestricted?
@JustSayin24
2 жыл бұрын
Actually the original research paper for DALL-E 2 states that text rendering is a known limitation of the model. Specifically, the "embedding does not precisely encode spelling information of rendered text" - in other words, the model isn't trained at a high-enough precision to properply represent the intracacies of charecter shapes and grammatical rules.
@tissuepaper9962
2 жыл бұрын
@@JustSayin24 I imagine they are already working on making the next model recognize text in the training data, transcribe it, and run it through a separate NLP model so that the image generator can understand grammar and spelling and stuff.
@cem_kaya
2 жыл бұрын
@@tissuepaper9962 there is no need to do such convoluted stuff to get photorealistic text generation. scaling up the model works fine
@mauricepouly
2 жыл бұрын
i adore your videos and the flair you bring into them. i enjoyed this one a lot too and it made me laugh which is a feat on its own. thank you phillip keep on doing what you do
@kruchji
2 жыл бұрын
I love how you immediately answer any question that I think of while watching. Great video!
@raminasadollahzadeh328
2 жыл бұрын
I am starting to get back to your channels. Been gone for about 2 years and now that I'm back I have to say there is a style in your videos which is very rare and unique. People say CSGO utubers are dying but you are proving them wrong. I am happy for you and for my self to find my new old fav channel.
@medelleude2cool
5 ай бұрын
It's hilarious seeing AI tools from just 2 years ago and how fast AI generated contend has evolved. Looking back at the now primitive DALL-E 2 is fascinating.
@fizzypizzel6477
Ай бұрын
ikr!
@Lulzalex
2 жыл бұрын
First the silent zoom in on the horse shaped entity followed by doing the same to the generated Chucky-doll all completely threw me off LMAO. I kept checking my back for the remainder of the video and could not relax like I usually do...
@boofkoosher2631
2 жыл бұрын
I was very shocked with dall-e 2 results. They were scaring-ly accurate and very detailed.
@Periwinkleaccount
2 жыл бұрын
Scarily.
@boofkoosher2631
2 жыл бұрын
Thank you dear sir for providing an appropriate word to fixate my lingua franca
@adan7949
2 жыл бұрын
It's actually a bit scary how convincing some of the images are, I'm glad those things are hard to get your hands on
@declanlambert1089
2 жыл бұрын
not for long
@BombBird11
2 жыл бұрын
@@happygofishing Dangerous little man, now aren't we? lol
@gigabooga
2 жыл бұрын
@@happygofishing Yeah but no one asked
@spiderjerusalem8505
2 жыл бұрын
@@happygofishing, true
@ShawnFumo
2 жыл бұрын
MidJourney has been letting many people in lately. Even DALL-E 2 said they let in 10k people in a week recently and had a survey on pricing models. It won’t be long.
@arcadianpunk
2 жыл бұрын
Let the battle begin
@WatamelonUberSheep
2 жыл бұрын
Son: Mom I want an EVA Mom: We already have an EVA at home sweetie Eva at home: 1:54
@CrazyKosai
2 жыл бұрын
more shenanigans with DALL-E 2 plz
@Sgt_Recka
2 жыл бұрын
I know you said that people don’t watch this kind of content from you. But I just wanted to say that I love it! AI is so interesting, and not many people on KZitem are showing what you are showing. I’m here for all your content, from all 3 channels
@DangOldRegularOld
2 жыл бұрын
12:36 original smiledog photo
@Krzeszny95
2 жыл бұрын
Wow, I wouldn't expect Dall-E 2 generating real faces
@DiveTheseClips
2 жыл бұрын
14:36 impressive result, true, but man I feel like those fellas are coming straight from uncanny valley
@Supreme_Lobster
2 жыл бұрын
We are witnessing the comodification of the creative process
@okolepuka808
2 жыл бұрын
Makes me want to take an entire descriptive paragraph from a book, see if it works.
@jeremywp123
2 жыл бұрын
This is quite dangerous, at least they're trying to take security somewhat seriously.
@Atreyuwu
2 жыл бұрын
I really enjoy AI generated art, just got my hands on Midjourney and installed local version of DiscoDiffusion (ProgRockDiffusion) and I'm loving it so far. What I don't enjoy is people who get me to watch their videos with a clickbaity title, and then they don't even really show me what the title implied lol. XD
@jamesfenimore174
2 жыл бұрын
The new title improved my interactivity.
@litterbox2010
2 жыл бұрын
It's creepy that DALLE 2 images look exactly how things look like in my dreams. Almost accurate, but not quite.
@saoliath5000
2 жыл бұрын
the tos on Dall-e 2 feels like its holding it back
@cURLybOi
2 жыл бұрын
the mattress is STANDING against the wall
@maymayman0
2 жыл бұрын
Sooo weird to me that almost all of the purple jumpsuits have green on them, like its got to be just because of the green hills being specified, it's so strange though how it always results in a purple and green jumpsuit
@vgaggia
2 жыл бұрын
You should try latent majesty diffusion and also don't forget to try using the --hd argument with midjourney.
@boggybolt6782
2 жыл бұрын
Insane that AI can generate an image that looks like a real photograph. I just wish the performance was better
@catcatcatcatcatcatcatcatcatca
2 жыл бұрын
It’s absolutely incredible. I think online article images (that aren’t about something specific that is discussed), moodboards and concept art for creative industry, and stock photos are soon gearing up for a complete revolution. I have no idea if and how our society can handle the change in nature of information, but then again highly realistic fake pictures have been possible for a long while. These will be widely available. At least for companies that are willing to pay.
@amnesicpachyderm
2 жыл бұрын
I'm loving these AI videos. It's an exciting and worrying time, and it feels like we're on the precipice of some historic developments. Hopefully good ones. But I guess we'll see either way.
@cobraofearth
2 жыл бұрын
The photos are already beyond what I thought was possible in my lifetime, but imagine AI that could construct video game environments to a photorealistic level. Could really streamline the process and get games out more quickly.
@phntm5700
2 жыл бұрын
this implies so much about the future
@rumplyscamp3700
2 жыл бұрын
You ever seen the effects of too much acid? Well now you have.
@SteveDave
2 жыл бұрын
as much as dall-e 2 blows my mind, i cant help but love midjourney's style. having spend hours trawling through the discord server, its like truly seeing dreams in reality.
@gtPacheko
2 жыл бұрын
Great videl from 26PKL!
@alexlokanin3312
2 жыл бұрын
this is so good for level design
@loetwiek
2 жыл бұрын
i love the ai things on your channel keep em coming
@Peztllence
2 жыл бұрын
I like Midjourney a lot. It carries a certain peculiarity to it that really sells the "this was made by an AI" feeling.
@Peztllence
2 жыл бұрын
It also carries in built grim reminders to not toy with it lest you return a changed man.
@FPSRayn
2 жыл бұрын
It would be interesting to see how the DALL-E Mini memes like "Bottle of ranch testifying in court" would look in each of these AI image generators.
@ocsanik502
2 жыл бұрын
Please generate photos of locations in the syle of the quake engine.
@michuXYZ
2 жыл бұрын
You can literally use it for creating textures
@antonholmberg911
2 жыл бұрын
i have gotten some almost photo real results in dalle mini like godzilla riding a unicycle (fisheye)
@containercore6832
2 жыл бұрын
3:17 this confirms that the images used aren't licensed in any way. I wonder if this is fair use or if there will be legal issues with it as more of this work starts getting used commercially.
@turke765
2 жыл бұрын
midjourney loves symmetry
@PUPIMUMIP
2 жыл бұрын
THIS VIDEO SHOULD BE SEEN BY THE WHOLE WORLD
@TheBestElectricToaster
2 жыл бұрын
One of the things I love doing with AI Image generators is having them make a hand and see how many have an incorrect amount of fingers. Another thing is "Man falls down stairs"
@Ogaitnas900
2 жыл бұрын
2klikphilip's bedroom is the ultimate captcha
@seruppo4219
2 жыл бұрын
Thank you for this video klik, have a wonderful day or and night.
@jholotanbest2688
2 жыл бұрын
This is the stuff everyone needs to know about.
@landonhagan450
2 жыл бұрын
Couldn't we theoretically train an AI to distinguish between real and AI generated images that are too convincing for the human eye?
@zubinzuro
2 жыл бұрын
That is what GANs (General adversarial networks) are for. They are trained to distinguish. However, many forms of AI are trained against GAN AI to become better and possibly eventually indistinguishable if that is the goal.
@Vanessa80808
2 жыл бұрын
Cant wait for when in the future ai will literally be able to generate animated shows where you could just feed it an entire book script and it will just make it into a show.
@everthealtruist
2 жыл бұрын
I make DALL-E make abstract stuff. Stuff from dreams, visions, and whatnot, like biblically accurate angels and stuff.
@FlyingWithFeathers
2 жыл бұрын
Did you just post DALL-E 2 real life images at the end tsk tsk
@lonergothonline
2 жыл бұрын
no testing different colour? like sepia or black and white? I wondered if it would do a better job of it if it only had to deal with greyscale, oh and pixelart.
@amp4105
2 жыл бұрын
imagine ai generated movies in the future
@nateg876
2 жыл бұрын
As someone who has the Midjourney AI from the discord, I have been astounded in the quality and exactly what I needed every time. I’m not sure what happened with yours, but to be fair I did sign up for Dalle 2 not midjourney. Are they the same? Would love a response before I pay for the subscription
@bjk0norway0bjk
2 жыл бұрын
really enjoyed this video :D
@JoRoBoYo
2 жыл бұрын
a.i is getting creepy without getting sentient at all.. lmao
@TheSliderW
Жыл бұрын
You should check out Stable diffusion. Even grt in running locally on your new GPU, that would be a treat. :)
@SpySappingMyKeyboard
2 жыл бұрын
I love the cursed images
@LetsGenocide
2 жыл бұрын
36KPK is my favorite youtuber
@pajokamikaze
2 жыл бұрын
I tried ''A shirtless buff man with spiky blonde hair in orange pants with a golden aura and electricity'' Anyone got the reference? c:
@youcantbeatk7006
2 жыл бұрын
Why didn't he just say "sideways mattress"
@gamecuber6
2 жыл бұрын
Imagen is sooo cool tbh
@cyjanek7818
2 жыл бұрын
8:40 maybe if you would specify it is vertical? Because it could assume it is close enough because it has to be horizontal and Thats the twist - it isnt.
@DavidOMPiano
2 жыл бұрын
What program(s) were used to upscale the AI generated images?
@BornAgainstAll
2 жыл бұрын
Not related to image generation/upscaling AI, but I highly suggest you pay attention to what the KZitem's AI recommends you if you really want to see how far AI technology has already come. It knows you better than you know yourself.
@kokki1452
2 жыл бұрын
I didn't imagine a mattress leaning up against the side of a wall - a mattress leaning against the wall on the left (the. mattress. is. leaning up. against. the. wall) - could make me lol so hard
@supremesurvivor
2 жыл бұрын
This is certainly one of my favorite videos on youtube, but it's so scary, unnerving, that I can't even describe what I'm feeling at the end. The feeling that we cannot predict what this might imply for art and politics without being pessimistic. Please Philip, keep it up!
@user-lh7mt7zo7l
2 жыл бұрын
It just means we'd return to a time before photo and video evidence.
@danisob3633
2 жыл бұрын
ye, lie detection needs to get better
@user-lh7mt7zo7l
2 жыл бұрын
@Lucas Carvalho I wonder what happens when we make AI generated images of people who don't exist but then someone is born who looks like that haha
@pygmalion8952
2 жыл бұрын
@@user-lh7mt7zo7l every ai service would be regulated to indicate it is an ai image in the photo's information. tho it is a bit shaky given the fact that you can spoof identification codes sometimes.
@user-lh7mt7zo7l
2 жыл бұрын
@@pygmalion8952 yeah regulation wouldn't work because with enough money and power you could make your own A.I.
@distortedjams
2 жыл бұрын
An interesting test would be to take a real life image, put that into a AI that can transcribe images (Instagram automatically does this). Feed that transcription to one of these AIs and compare the results.
@xouthpaw
2 жыл бұрын
And then Instagram won't need any content creators anymore, because you'll be able to log in and receive a 100% AI generated image feed based on your assumed preferences
@IronKurone
2 жыл бұрын
@@xouthpaw Few years later from today, perhaps my favorite instagram celebrity might not even be human. And that...kinda scary.
@treudden
2 жыл бұрын
You can use an init image in disco diffusion which works really good
@Erveon
2 жыл бұрын
@@IronKurone Knowing people have a favorite instagram celebrity is by itself scary enough
@IronKurone
2 жыл бұрын
@@Erveon Its the future, who know...
@Zoo-Wee-Mama-Sq
2 жыл бұрын
It's been a joy watching your channel branch out from CSGO mapping topics to technology in general, while still bringing the same top notch production.
@vankata69exe45
2 жыл бұрын
philip is an ai with very good text to speech at this point
@existentialselkath1264
2 жыл бұрын
New York in unreal engine is genuinely really impressive. It doesn't just look like a game, its got that distinct unreal engine 4 look I can never explain, but it's done it perfectly
@arrowtongue
2 жыл бұрын
AI generated images are so great at capturing the feel or vibe of something, because the nature of neural networks is stuff we can't quite describe, it's as scary as it is weirdly comforting we can turn these more abstract feelings into things
@ChrisD__
2 жыл бұрын
I think it's the orange sun like paired with blue everything else, the artist's color grading goals leaking into the actual world lighting. Along with missing shadows here and there and repeated objects and textures. Notice all the fire escapes all over the place. Also the general blurriness of the bounce lighting.
@Strelokos666
2 жыл бұрын
"distinct unreal engine 4 look" what the hell is that suppose to mean?
@ChrisD__
2 жыл бұрын
@@Strelokos666 Ya know... that UE4 look. Orange and teal, TAA, dithering, and every post processing effect under the sun.
@eldarlrd
2 жыл бұрын
@@Strelokos666 You haven't played any UE4 game?
@arrowtongue
2 жыл бұрын
8:40 your disappointment with the mattresses and stubborn valve please fix made me genuinely burst out laughing, love your sense of humor
@DeepWeeb
2 жыл бұрын
Petiton to rename the channel to *"3klikspiphlipk"*
@Snowdrift72
2 жыл бұрын
8:08 is the body the AI has created for itself and chosen to inhabit
@olegmoki
2 жыл бұрын
If you use DALL-E at 3 am and then turn around... ᅠ
@1000_Gibibit
2 жыл бұрын
Really glad that you managed to get (direct or indirect) access to DALL-E 2. These comparisons are wonderful! And of course you came up with some great prompts for the AI as always. The rate at which AI research advances is actually insane. And the conditions required for this pace, like rapidly improving hardware, are starting to feel like they are straight out of a sci fi story if you think about it. How long before someone accidentally creates an AI that can operate on real life systems that we lose control over? I always thought AI doomsday thinkers were too optimistic about AI. Now I don't know anymore if it's possible for a story like Hyperion to become reality. All bets are off. Oh and all the shorter term consequences relating to reliability of image validity are getting a bit concerning as well of course...
@oldm9228
2 жыл бұрын
GitHub copilot is probably an example of a currently active AI that operates on real life systems. It generates context aware code for applications based on requests. The quality of that code is questionable and it could potentially include "hidden intentions" (security risks) just like human written code can.
@HighWarlordJC
2 жыл бұрын
There's a very real reason many of our brightest minds constantly warn about the dangers of AI.
@amp4105
2 жыл бұрын
imagine ai generated movies
@amunak_
2 жыл бұрын
@@oldm9228 Copilot and similar are amazing for generating boilerplate and small chunks of code that you can actually verify yourself. But I have doubts about usage beyond that.
@McDonaldsCalifornia
2 жыл бұрын
I mean dall-e and gpt and stuff are genuinely impressive but they are far from what we would expect a true AI (or AGI or Super AI or whatever) to look like.
@nixel1324
2 жыл бұрын
Yes, Dall-e mini (now Craiyon) has a very ai-y feel to it, but I like that. It's like the charm of a retro console. From a technical standpoint it's inferior in every way, but that makes it recognizable, gives it character and makes it endearing. And once people grow up in a world where the higher end stuff is the norm, people like me will probably largely be considered old-fashioned. I don't really care much about modern consoles, and cannot tell apart PS5 and Xbox X footage. But I'll instantly recognize a Wii game, even if running in 4k with texture replacements. Even when you upscale Craiyon results (like with Dall-e Flow), it still has that charm for me. When photo-real ai images become mainstream, I hope people will still appreciate the weirder, less fine-tuned options. I think I will, at least.
@KVVUZRSCHK
2 жыл бұрын
Dall-E Mini is on the left side of the uncanny valley. Astonishingly lifelike yet easily distinguishable as fake.
@IndieLambda
2 жыл бұрын
That's when you add "AI generated" at the end of your prompt.
@RaptorShadow
2 жыл бұрын
Someone pointed out that the surreal and disposable quality of the Craiyon images make it perfect for memes. You can quickly and cheaply get a rendering of whatever stupid idea you come up with (like Boris Johnson's Bean Bath Suprise) and get some output. The jank becomes part of the charm.
@deKxiAU
2 жыл бұрын
Just a tip for 'photorealism' with these models: put camera / photography specs like F-stop, iso and lens length - works best with DALLE 2 Edit: ah I see you've done that in the later prompts, ignore the above then :P Also worth noting none of the generation methods actually merge images from Google, they just had watermarked images in the dataset. I realise you probably don't think that it actually does just google some images given what you said, and it might seem pedantic - but it's a proper distinction (and the 'just slapping images from googling together' myth is a very common for all AI generative art right now), for those who don't know: the model's learned that that particular image is likely to have a watermark from what its seen in its training dataset and so it's synthesised it. It's not actually searching anything on any search engine or anything like that, it's just a matter of the dataset not being cleaned of any watermarked content Great video though Philip :) Edit: also for Midjourney specifically, there's some additional background style modifiers you can disable that would be somewhat influencing your result out of the box for the cartoon ones and making them less accurate to the prompt, forgot what the arg is as I'm on mobile watching this but it's somewhere in the FAQ I believe - but this is why you always get a vignette, a similar colour palette, amongst others behaviour across every Midjourney prompt
@huttyblue
2 жыл бұрын
What was a watermarked image doing in its data set if it wasn't sourced from scraping the web though? It may not be specifically from google but the concept of it just learning from what was able to be searched up on the internet is the same.
@deKxiAU
2 жыл бұрын
@@huttyblue I didn't say it wasn't due to web scraping, it absolutely is. I'm saying it's not googling/searching online at inference stage (the stage where people can actually interact with the AI in the way you see in the video), and in fact the AI never touches the internet (outside of it getting hosted online to be accessed by yourself, or it being trained using a GPU cloud farm somewhere in the first place). It's a large fundamentally different procedure with very different outcomes. Since most people here probably aren't familiar with neural net training I'll elaborate a bit: CLIP was trained on web scraped images as outlined above (CLIP being the model under the hood of most AI generative apps/notebooks and of Midjourney too), but it's nowhere near the same as a program searching up your prompt for close images online that match and then splicing those together - it's not a glorified Pinterest board. The dataset is static from the date of when it was scraped and published. It's then used as training material for the basis of the AI's generative ability - you won't find things posted after the date of the dataset in it's generative vocabulary for example. Naturally, a poor dataset can lead to poor results and watermarks are an obvious poor side effect of web scraping, but to conflate it with 'searching online' gives the impression that it's simply reverse searching for your images and slapping them together which leads to people believing that it is actively searching and essentially 'cheating' - like someone looking up results for a test right? Whereas in actuality it genuinely generates the images based on what it learned from 'studying' the dataset and associating different labels with what it thinks is relevant, as if it spent it's time studying wikipedia articles instead of the sources wikipedia lists, etc. CLIP has stupidly learned that stock image watermarks are common enough across it's whole dataset that they are worth adding to some images sometimes even when not directly prompted for it, because it had enough images to train on that had watermarks that it associated the concept of watermarks with that sort of image in its latent space. But it's the concepts themselves that it has learned, not direct image portions and mashing them together. DALL-E 2 has this same issue but the dataset was far more curated, it's fairly difficult to get a blatant example. DALL-E Mini (now CrAIyon) also suffers from this but the quality is bad enough that you'll be hard pressed to even recognise it's a watermark and not just random jibberish text. Most models at the moment are trained with the LAION dataset (among a few more) which has a whole host of web scraped content (including graphic porn and all sorts of NSFW images - these usually get taken out manually by the big companies models), but until there are open sourced datasets that don't have to rely on web scraping to get the sheer number of images training a model requires (several hundred million to billions), stuff like watermarks and weird quirks are just part of the parcel - that said, web scraping is also why it can make such hilarious memes because the highly curated datasets (like the one in DALL-E 2) remove large chunks of the image base and sort of gut the models ability to accurately reach a prompt in the process. TLDR: Its the difference between studying for a test before the day, or actively searching online during your test. Hope that helped illuminate the differences! Enjoy your day :)
@tissuepaper9962
2 жыл бұрын
@@deKxiAU I disagree that there's much of a difference. The claim that it's "just slapping images together" is basically pointing out that the system doesn't know anything about *why* certain features exist in images, it just knows that they *do*. AI at this point are still just advanced statistical aggregators, most lack the kind of logic that would allow them to generate images with details that make sense as opposed to just looking right at a glance. Philip isn't saying that the system literally merges images from Google at the time of inference, it seems to me like a subtle statement about "learning" vs. "regurgitating" and what should actually be called "intelligence".
@deKxiAU
2 жыл бұрын
@@tissuepaper9962 there is a significant difference. If Philip meant that 'it doesn't know why features exist' he should have just said that. Learning the 'wrong' details doesn't make it 'regurgitation' any more than learning the right details would, and it falling apart under scrutiny is largely due to the limited resolution of the training data (typically 256x256 or 512x512, DALL-E 2 starts at 64x64 with additional diffusion networks trained on upscaling it incrementally) combined with the limited number of parameters the model contains which leads to it having to combine concepts into the same latent dimensions and differentiate between them poorly as a result. I'm not sure what you could disagree with really, like I said at the bottom - it's the difference between studying for a test or looking up the answers during it, entirely different implications can be drawn from systems that do either of those. The former relies on prelearning concepts and identifying key relationships between them, the latter can pick new images as they pop up on the internet and doesn't have any understanding of the relationship between concepts at all. One is learning conceptual relationships, the other is a pinterest board with a fancy text input. I'm not saying it's not statistical aggregation, I'm just saying it's not ripping images off the internet and splicing them together like some Frankenstein creation, and that it *has* learned within the weights of its millions to billions of parameters that there is an association between watermarks and those types of images - which is actually true, in the dataset it was trained on there was enough watermarks for it to recognise the concept across them and learn about it the same way it has for every other concept it recognised; like trees and bushes belonging in a forest, stock watermarks belong on stock-looking images. Removing the watermarks from the dataset would solve that specific issue, but wouldn't change anything about how its fundamentally working, it would just give better results because it's an algorithm that aims to be able to create images that *could* have been from its dataset without actually recreating any image from it (that would be what's called overfitting, which we dont see in these models), its task is quite literally to map the entire range of possible images in its dataset and to abstract whatever relationships it can to condense it into its embedded parameter weights, and so it would be a failure if it didn't have watermarks when there are so many in the dataset. Make sense? Intelligence has nothing to do with it, different conversation entirely. Not arguing it's sentient or that it understands in a way that human brains do, (obviously, the way it understands and learns isn't as complex as humans and it doesn't have an understanding of *why* these things exist together, just that they do, because the why wasn't part of the training data - its simply condensing image concept relationships to an extremely large matrix of numbers), just that people shouldn't propagate a myth because "it's close enough" when it actually gives a false impression of what these models can do and how they work, and what it means for the world; different behaviour, different results, different legal implications, different world outcomes and use-cases. I'm only hoping to help correct the record as I'm a huge fan of Philips content - not wanting to knock the video, overall it's very good and knowledgeable and at the incredibly high standard Philip always provides for his videos - just that particular line (which he said twice) suggests hes either a bit misinformed on the topic (which is fine, everyone's misinformed on something and it shouldn't be taken personally if it's corrected) or that he wasn't quite clear on what he meant (also fine as he possibly wasn't aware of how what he said could be interpreted)
@tissuepaper9962
2 жыл бұрын
@@deKxiAU You have your interpretation, I have mine. You can carefully defend the model by explaining the limitations, that doesn't change my opinion. I think it's a perfectly acceptable simplification made for brevity, something you appear to hold in little regard. PS: You say "intelligence" is a different discussion, did you forget what "AI" stands for?
@simian.friends
2 жыл бұрын
your writing and presentation is particularly great in this video, really enjoyed this, already can tell that I will be rewatching this many times over the coming months
@mattd1466
2 жыл бұрын
I'm not sure you're aware of how good you are at presenting and making topics interesting, like I still watch your csgo videos even though I haven't been playing the game for years because they're still enjoyable to watch.
@mattd1466
2 жыл бұрын
@@2kliksphilip oh totally! at the end of the day I prefer my Philip kliked twice over thrice.
@JohnDoe-sw2nc
2 жыл бұрын
DALL-E 2 is scary good
@KrynexYT
2 жыл бұрын
As Károly from Two-Minute Papers always says, imagine the improvement two papers down the line. DALL-E 4 will probably make graphic designers etc. largely redundant.
@bluebell2334
2 жыл бұрын
I love Karoly's style of presenting something. Each video exceeds my expectations.
@s-zz
2 жыл бұрын
The irony of it all, is the fact that a lot of the same AI designers are also working on AI that can code. And will eventually cause them to become obsolete too. Seriously, look up coding with AI, there's a lot of info on it already.
@rene-of3sc
2 жыл бұрын
@@s-zz Meh, Copilot for example is useful to generate easy or repetitive functions but no matter what, a human would need to say what needs to be generated and see if the generated code is correct. I would assume in the future AI will be used as a productivity tool by programmers, but not replace them.
@AlphaGarg
2 жыл бұрын
@@rene-of3sc This. I hate this whole "[job] will become redundant!" falsity that people have for some reason hung onto. Did Photoshop make photographers' jobs redundant? No! Did node-based programming like Unreal's blueprints make programmers' jobs redundant? No! Neural networks like DALL-E, Jukebox, etc. are tools that'll be used by the people that know the most about these things - artists. Sure, any old schmuck might be able to generate an image based on a prompt. But they aren't going to be able to do it the same way an artist will. Artbreeder has existed for a while now, yet outside of artist circles, I haven't seen that much use of it. Same will happen to these once they get normalised and become accessible.
@trallakid
2 жыл бұрын
i don't think it will make all graphic designers redundant, just the ones stuck in the past. with all professions, the technology is constantly changing so any good graphic designers would ideally use these types of technology as another tool in the toolbelt. As a graphic designer myself I can 100% see this technology being great for idea generation and coming up with some ideas from prompts, but I don't think it will ever be able to fully replace a human (although mark my words I might regret going down this career path in a few years lol)
@hisshame
2 жыл бұрын
Thank you for sharing the process with us!
@MattVidPro
2 жыл бұрын
great video! I've been making a plethora of videos discussing and testing this technology lately, and man is it moving FAST. Every few days I hear something new....
@iulic9833
2 жыл бұрын
I know, can't wait for DALLE 2 to get released to the public, if it ever will. Also got some good results when upscaling the images too, they have some artifacts but its still mind boggling how an AI can create stuff as this. kzitem.info/news/bejne/wmeAmYilf4dyi20&ab_channel=69fff
@luna010
2 жыл бұрын
tbf, midjourney’s first result was definitely the most interesting, and I think it fulfilled the prompt. the dalle2 results look like shitty google images clipart.
@ipixz3
2 жыл бұрын
Wouldn't it make more sense for Concept Artists to be considered obsolete instead?
@luna010
2 жыл бұрын
I feel like the more realistic AI generated images become, the more people will appreciate how cool “bad” AI generated images can be. The novelty of photorealistic images being AI generated will wear off once it’s commonplace, but images that don’t look real will always be at least a little bit interesting.
@SBImNotWritingMyNameHere
2 жыл бұрын
put timestamp so more people get what youre talking about pls
@TheKrzysiek
2 жыл бұрын
While others are worried about using this for more malicious stuff, I'm more excited about how much cool new content we can get from this. Want a specific image for a video, wallpaper, or a meme? Put it in AI I especially wonder if it will ever be used for things like concept art, book covers, character portraits etc.
@devindykstra
2 жыл бұрын
Of all the things to defeat Dalle 2, I would never have expected a mattress leaning against a wall.
@seto007
2 жыл бұрын
Hey Philip, I recently got access to both DALL-E 2 and Midjourney, and so I wanted to share a bit of my perspective on the strengths and weaknesses of both. While DALL-E 2 is certainly better at generating the initial image at a higher fidelity and with more stylization based on the description provided, I actually think that Midjourney succeeds far more at creating a "final image" than DALL-E 2 does. The reason for this is that the subsequent variations of an image that you can generate with DALL-E 2 often deviate significantly from the original description, to the point where it often feels as though the AI is trying to guess at what the original description you used was based on the image it's making variations from, and because of this it often feels like the AI subsequently gets confused and creates more abstract renditions than what you might have intended. Midjourney doesn't seem to have this issue. Subsequent variations seem to stick to both the original description and intent behind the image it's creating variations of, and because of this it feels as though subsequent generations look much closer to the original intent of the person describing the image to be generated. Beyond this, it feels as though DALL-E 2 has some issues with understanding things like perspective in all but the most simple of circumstances. If you were to ask it to generate an image viewed from the side, for example, it will often give you an image viewed from a diagonal downwards angle, as opposed to a true sideshot like what you would see in something like a Shutterstock photo. Midjourney does not have this issue in most circumstances; it seems to understand that you want to view the object being described from a side-facing angle. I think both models have their strengths and weaknesses, depending on the use case; since I am primarily interested in using these AIs to speed up the art process for a cyberpunk video game I am working on, I like using DALL-E 2 to generate stylized concept art that gets across the themes I am going for, whilst I prefer using Midjourney to generate more technical images of hypothetical in-game objects to use as reference.
@lukasg4807
2 жыл бұрын
TBH I'm more impressed with the ability to understand what you're asking for than the image generation itself
@HELLF1RE9
2 жыл бұрын
8:06 that is unbelievably unnerving
@keenban
2 жыл бұрын
I just got access to Dall-E 2 the other day, and I have been playing around with it. Honestly, it is quite crazy what it can do. I wonder how it would be if it were unrestricted.
@BombBird11
2 жыл бұрын
*C H A O S .* Just pure, utter chaos....💀
@EmilySmirleGURPS
2 жыл бұрын
The reason the AIs all had trouble with your "Mattress leaning against the wall" has to do with training data. The AIs don't have generalized concepts like "This object can be rotated in 3d space freely and still be the same object." - each classification of objects needs to learn that idea separately. They're better at handling some things in different orientations (particularly animals and people) because we have given them lots of things labeled "cat" in all sorts of postures, therefore they know cats can be upside down or backwards or curled into balls or etc etc etc. My suspicion is that they didn't get pictures of mattresses in anything other than a horizontal position (it seems Dall-e mini also only got them as part of a bed!) therefore mattresses are things that are horizontal (and part of a bed, if you're Mini). The "Shutterstock" watermark, by the way, isn't because it stole the original image from Shutterstock. It's because it included a lot of images from Shutterstock in its training data - therefore it's learned that you can chuck the Shutterstock watermark on a *lot* of different images and be "valid", so it tries now and then. There's a couple of databases of images used to train AIs that have these Shutterstock-watermarked images in them, so it's a pretty common quirk of computer vision / visualization. These AIs are "creative" in the very literal sense of "creating" images, but you can ask a 4 year old to draw a car on its roof or standing on its bumper and even if they've never seen a car do this, they can imagine it and draw it crudely. An AI cannot. It needs to *explicitly* be told cars "can" do that and still be cars.
@artemisDev
2 жыл бұрын
the new Turing test: "Draw a mattress leaning up against wall".
@godofzombi
2 жыл бұрын
I've found Dall E mini is really good at Art Deco posters, especially if you stick to natural landscapes. H.R. Giger also gives decent results, altough not the best quality. And mini's tendency to mangle faces makes some drawings almost look like the works of Francis Bacon.
@MarkSulekTalk
2 жыл бұрын
For your interest, I've seen an article where a photographer integrated an image he took in DALL-E 2, which was blurry and out of focus, and writted "Ladybug on a leaf, focus stacked high resolution macro photograph". The image recovered details and focus and became tack sharp, which was impressive ! You should try to do that !
Пікірлер: 803