Deep fake technology is shockingly simple to use. I decided to make one to deceive my audience, and it worked.
Get access to behind-the-scenes vlogs, my scripts, and extended interviews over at https://www.patreon.com/johnnyharris
I made a poster about maps - check it out: https://store.dftba.com/products/all-maps-are-wrong-poster
Custom Presets & LUTs [what we use]: https://store.dftba.com/products/johnny-iz-luts-and-presets
The music for this video, created by our in house composer Tom Fox, is available on our music channel, The Music Room! Follow the link to hear this soundtrack and many more: [[[INSERT LINK]]]]
About:
Johnny Harris is an Emmy-winning independent journalist and contributor to the New York Times. Based in Washington, DC, Harris reports on interesting trends and stories domestically and around the globe, publishing to his audience of over 3.5 million on Youtube. Harris produced and hosted the twice Emmy-nominated series Borders for Vox Media. His visual style blends motion graphics with cinematic videography to create content that explains complex issues in relatable ways.
- press -
NYTimes: https://www.nytimes.com/2021/11/09/opinion/democrats-blue-states-legislation.html
NYTimes: https://www.nytimes.com/video/opinion/100000007358968/covid-pandemic-us-response.html
Vox Borders: https://www.youtube.com/watch?v=hLrFyjGZ9NU
NPR Planet Money: https://www.npr.org/transcripts/1072164745
- where to find me -
Instagram: https://www.instagram.com/johnny.harris/
Tiktok: https://www.tiktok.com/@johnny.harris
Facebook: https://www.facebook.com/JohnnyHarrisVox
Iz's (my wife's) channel: https://www.youtube.com/iz-harris
- how i make my videos -
Tom Fox makes my music, work with him here: https://tfbeats.com/
I make maps using this AE Plugin: https://aescripts.com/geolayers/?aff=77
All the gear I use: https://www.izharris.com/gear-guide
- my courses -
Learn a language: https://brighttrip.com/course/language/
Visual storytelling: https://www.brighttrip.com/courses/visual-storytelling
You can watch the full video here: https://youtu.be/S951cdansBI
@kelseybach02ify Says:
What happened to the full video of this? I saw it months ago and wanted to share it and the link above is no longer actvie
@naasvanrooyen2894 Says:
Deep fakes arnt specific to GANs per se. GANs are just a special way to train the neural network, and may or may not work well for specific problems, types of deep fakes. My point is, its like saying a BMW is allowing transportation. Cars are allowing it, a BMW is good for open roads but maybe not dirt roads for example. Nonetheless still love your videos!
@DAP_TECH Says:
What software did you use?
@BSharp369 Says:
Very scary and future implications of this are horrifying
@MCEdziits Says:
There is AI picture, AI audio and AI video. Soon there will be AI human body twin of you.
@jennessacard4796 Says:
It’s looks like it can be your son but not you
@eds1445 Says:
Wow that is like the equivalent of evolution for AI
@Opinions567 Says:
Looks like your brother but not you
@janinegaylard9233 Says:
As a graphic artist I recognise this toxic loop
@winglessepyon Says:
This certainly won't be abused in the future...
@cacofany1 Says:
It looks good, but the Ears are wrong 😂 it’s actually funny it looks like the ears are slowly pealing back from the camera
@wapper7777 Says:
AI should be banned and create AI should be suitable to capital punishment
@ricobernaldo9969 Says:
Let's all agree that the deepfake he created does not look like him 😊
@MrCoolprax Says:
Developer vs tester
@TempestIsa Says:
Amazing, and incredibly disturbing!
@happilyevernever4289 Says:
Im never showing my face ever again.
@1337lll Says:
Nick needs to fix his mustache 😂😂😂😂😂😂😂😂😂😂😂😂😂😂
@tristan7216 Says:
Why don't we use an adversarial "verifier" AI to tell how far from verifiable facts an AI's output is, so we don't get AI making up fake legal case law and similar in its output?
@sarahmurphy-nf4yl Says:
SCARY 😨
our iPhone and android phones take photos of us every 5 seconds it flashes but it can only be seen by infra red camera!!! Are they taking these images of EVERYONE to use for this or to PUT OUR FACES ON AI 😨 😱
@CaseyMusicOfficial Says:
Do you know how to make it useles? To not watch it 😅😅😅😅😅
@antonydandrea Says:
Dont know who fell for that
@michaelpettersson4919 Says:
That "fake" look like a younger version of yourself. Hollywood want to do this to de age their stars. They could also let someone else play the role and then transform this into looking and sounding like their preferred star even decades after said person passed away. We could get brand new movies "starring" Humphrey Bogart. What his surviving relatives think about that I do not know but I imagine that attempts to buy them off could be done. And this would just be a temporary problem anyway, at some point any arising star could be given eternity contracts where the company owns the right to keep using their likeness and voices indefinitely.
@carolineapodaca525 Says:
What surprises me more is we're seeing politicians and their heavily devoted followers already claiming any evidence found against them are deep fakes then we are actually finding deep fakes.
@tedmusson5179 Says:
Does anyone hear anything worth hearing from our boy here. I don't.
@tg7048 Says:
AI is going to throw us into dystopia instead of an advanced civilization. My reasoning is because of how terrible and stupid humans are still. I can see it being useful for an advanced society, but we’re no where on this planet close to that. Culture wars, religious wars, political wars, racism, etc….its all so stupid. Maybe I’m wrong in my reasoning though.🤷🏻♀️
@jondoe2542 Says:
Basically how I explain the reason Joe Biden is actually a robot.
@miyamotomusashi5480 Says:
From GANS to GANTS when???
@Burningarrow7 Says:
AI will be the end of our society as we know it
@Yourname-d1k Says:
Lets get back to button phones
Its the only way with some messaging apps
@Papa_Tea Says:
If you fall for that you have the obvservation skills of a piece of cardboard.
@buzzkillington1719 Says:
Didn’t deceive me
You are handsome
Don’t bring a nerdy looking guy into this
@barryhopkins4765 Says:
Watch the 1970 movie; Colossus about two A,I computers and this will happen. A,I is dangerous.
@Phillious190 Says:
Scary stuff
@folklorian4368 Says:
Wait, that's not have GAN's work
@sweeperbot9747 Says:
GANs aren’t nearly as effective as stable diffusion. The problem with GANs is they can undergo mode collapse where the generator finds one particularly good example and only generates that as the output. The detector then never learns and the generator never improves. This can be corrected with Wasserstein loss, which calculates the loss based on the distribution of the whole dataset rather than each individual piece of data generated, but in either case stable diffusion performs much better
@Packee Says:
This technology it's the beginning of the end of reality.
@jvh4438 Says:
Lmao why do I get liberal propaganda channel #5470 in my shorts feed.
@Mahteye Says:
The only question remains: Who is Nick the studio manager?
@seymaple Says:
You should do mcgreber, and do his grace yard scene
@TheShockingShane Says:
Deep fakes are the governments golden child. They absolutely love that this is a thing. Now they can deny any and everything again. Like they could years ago. The next democratic candidate can go on and just scream the N word infront of thousands. And they can just say it's fake and that it's a deep fake. Easy.
@pablogarcia6188 Says:
I dislike how present information even tho i like the topics you discusse
@memesfamilyguyandtvshows Says:
The end of humanity is here and am loving it,😂😂
@strategistaow3520 Says:
This is why a.i needs stop
@ericwalsh2954 Says:
The lips don't move adequately for the sounds the make, otherwise pretty good
@wiktorpp Says:
The adversary doesn't tell the model what went wrong. It just guesses if the image is from the dataset of created by the model. Am I wrong? Brcause what he said sounds very wrong to me.
Edit:
What's the gradient in an image GAN?
Bing chat replied:
In a GAN, the gradient is the direction of the steepest increase in the loss function. The generator tries to minimize this loss function while the discriminator tries to maximize it. The gradient is used to update the weights of the generator and discriminator networks during training¹.
In MSG-GAN (Multi-Scale Gradient Generative Adversarial Network), gradients are allowed to flow from the discriminator to the generator at multiple scales. This technique provides a stable approach for high resolution image synthesis, and serves as an alternative to the commonly used progressive growing technique¹.
@golddropper2747 Says:
You look like a mix of pewdiepie and oompaville
@mustard_owl5650 Says:
You dont need a deepfake to that
@Mayra_mayhem4247 Says:
WE BEEN AI THE WHOLE TIME!!!!!
@SurajKumar-ln8ij Says:
The finer the technology getting the more wr getting to close to simulate the reality absolute.
LATEST COMMENTS