Jump to content

Welcome to FutureTimeline.forum
Register now to gain access to all of our features. Once registered and logged in, you will be able to create topics, post replies to existing threads, give reputation to your fellow members, get your own private messenger, post status updates, manage your profile and so much more. If you already have an account, login here - otherwise create an account for free today!
Photo

Synthetic Media & Deepfakes News and Discussions

Artificial intelligence synthetic media deepfakes media synthesis GANs artificial imagination image synthesis natural language generation automation deep learning

  • Please log in to reply
298 replies to this topic

Poll: Synthetic Media Poll (25 member(s) have cast votes)

When will deepfakes be perfected?

  1. 2020 to 2022 (5 votes [20.00%])

    Percentage of vote: 20.00%

  2. 2022 to 2024 (8 votes [32.00%])

    Percentage of vote: 32.00%

  3. 2024 to 2026 (5 votes [20.00%])

    Percentage of vote: 20.00%

  4. After 2026 (7 votes [28.00%])

    Percentage of vote: 28.00%

Vote Guests cannot vote

#281
crusher

crusher

    Member

  • Members
  • PipPip
  • 21 posts

This Tom Cruise TikTok stuff is absolutely fucking insane. I saw it on TikTok without context and was fooled into thinking that it's really him even though I know about deepfakes and have seen a ton of them. I think I just had a futureshock moment.

 

I see now that blackbomb posted a twitter link to one of those tiktoks but I'll post this additional link anyway cause it contains all 3 videos and a little commentary, so here it is:

 

https://www.independ...o-b1808000.html



#282
Miky617

Miky617

    Member

  • Members
  • PipPipPip
  • 59 posts
  • LocationNC

This Tom Cruise TikTok stuff is absolutely fucking insane. I saw it on TikTok without context and was fooled into thinking that it's really him even though I know about deepfakes and have seen a ton of them. I think I just had a futureshock moment.

 

I see now that blackbomb posted a twitter link to one of those tiktoks but I'll post this additional link anyway cause it contains all 3 videos and a little commentary, so here it is:

 

https://www.independ...o-b1808000.html

 

I saw two of those last week and I'll admit, it's almost freakish how realistic they were. If I didn't already know ahead of time they were DeepFakes, I would have been fooled. We're starting to climb out of the Uncanny Valley with them and honestly I'm amazed that we haven't yet seen them become widely used for political purposes. Though some might say as convincing as they're getting, perhaps we've already encountered a few in the wild and just don't know it. 

 

Part of me wants to be optimistic and hope that because of this, people will become more skeptical and critical of information they come across online. But then again... 



#283
Raklian

Raklian

    An Immortal In The Making

  • Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 7,517 posts
  • LocationRaleigh, NC

We'll have to use every ounce of intelligence we have not to be tricked by deepfakes. 

 

Having said that, I feel a sense of dread for Americans at large given they were easily tricked to vote for a conman.


What are you without the sum of your parts?

#284
Yuli Ban

Yuli Ban

    Born Again Singularitarian

  • Moderators
  • PipPipPipPipPipPipPipPipPipPipPip
  • 22,999 posts
  • LocationNew Orleans, LA

^ The existence of QAnon shattered any and all lingering doubts I may have had about the propensity of people to realize that digital media can lie even if it says what they want it to say. 

The cold fact is that communities will have media tailored specifically to them and what they want to hear, keeping them from leaving. No matter where you are or what you've learned, this is nothing like what we're used to. If no one told you this Tom Cruise deepfake was a deepfake, would you have thought it was a deepfake? The only reason we're figuring it out this easy right now is because the tech is still deeply immature and because everyone loudly announces "Hey, look at this supercool deepfake!" every time. The first time someone doesn't announce it and the product is good enough to pass for reality is all that matters.

 

 

The danger is thinking that synthetic media can't go that far. That it'll reach some asymptote of almost seeming realistic but falling just slightly short, and only affects talking heads and maybe cheap news articles. If that's what you think will be the limit of synthetic media in the near future, you might as well put on clown makeup now.


And remember my friend, future events such as these will affect you in the future.


#285
funkervogt

funkervogt

    Member

  • Members
  • PipPipPipPipPipPipPip
  • 1,323 posts

This Tom Cruise TikTok stuff is absolutely fucking insane. I saw it on TikTok without context and was fooled into thinking that it's really him even though I know about deepfakes and have seen a ton of them. I think I just had a futureshock moment.

 

I see now that blackbomb posted a twitter link to one of those tiktoks but I'll post this additional link anyway cause it contains all 3 videos and a little commentary, so here it is:

 

https://www.independ...o-b1808000.html

 

I just watched it. Insane. The technology has advanced faster than I expected. 

 

The Deepfake didn't just reproduce his face--it accurately duplicated his body mannerisms. Yes, I know that was due to the human performer doing his own research, but it makes me realize that aspect of individual human behavior can also be quantified and reproduced by machines. 

 

 

^ The existence of QAnon shattered any and all lingering doubts I may have had about the propensity of people to realize that digital media can lie even if it says what they want it to say. 

The cold fact is that communities will have media tailored specifically to them and what they want to hear, keeping them from leaving. No matter where you are or what you've learned, this is nothing like what we're used to. If no one told you this Tom Cruise deepfake was a deepfake, would you have thought it was a deepfake? 

 

I predict that, in the not-too-distant-future, deepfake videos of a famous dead person will be shared online by millions of people, leading to an internet-based conspiracy theory that the person is alive and merely faked his or her death to "get off the grid" or whatever. The deepfakes will be indistinguishable from real footage, no one will know who is making them (imagine the "Q Drops" or whatever), they will have a consistent agenda (i.e. - cultural or political), and eventually, the persistence of the theory will lead to serious calls to dig up the dead person's grave to verify whether their corpse is there. 

 

Come to think of it, this would work best if the dead person had been cremated. 



#286
caltrek

caltrek

    Member

  • Members
  • PipPipPipPipPipPipPipPipPipPipPip
  • 12,797 posts

I think that there will be  at least an interim period where experts will still be able to determine whether a video is a deep fake or of the real thing. The problem may be who will believe such experts?

 

I mean if there are still people that:

 

  1. Doubt that global climate change has anthropogenic causes despite what experts tell them.
  2. Think that the 2020 election was stolen despite what experts tell them.
  3.  Believe that the case is unproven that smoking leads to many getting cancer.
  4. Think that government operatives are some how in touch with extraterrestrial intelligent aliens and just have not gotten around to telling the rest of us.
  5. Think we are living in a simulation.

If there is this level of irrationality and/or distrust, then surely deep fakes may very well end up adding to the confusion.

 

A majority may be persuaded that a given high quality deep fake is not genuine, but what about that minority who do not get or do not believe the memo?


The principles of justice define an appropriate path between dogmatism and intolerance on the one side, and a reductionism which regards religion and morality as mere preferences on the other.   - John Rawls


#287
rennerpetey

rennerpetey

    Member

  • Members
  • PipPipPipPipPipPip
  • 605 posts

 

Think we are living in a simulation.

 

 

On this one point I'd posit that there's no way to know if we are or not.  If you believe humanity will eventually be able to simulate reality then you also must accept that there is a chance we are in that simulated reality.  That said, to us it makes no difference because the simulation IS our reality.


Love and let live


#288
caltrek

caltrek

    Member

  • Members
  • PipPipPipPipPipPipPipPipPipPipPip
  • 12,797 posts

+1 for Rennerpetey.

 

The question of simulation brings up the question who (or what) is doing the simulation?

 

For some, the answer may be "God."   

 

A subdivision of those with such an answer are those who do not necessarily give up on ideas such as the scientific method or love.  Both are ideas that can be applied to "our reality."  

 

Many would respond negatively to this, possibly because of unpleasant encounters with conservative Christians.  At the risk of derailing this particular thread, here is an interesting article I just came across which argues that there is, at least in the United States, a resurgence in the religious left:

 

You Need to Take the Religious Left Seriously This Time

 

Introduction:

(Politico) For generations, American politics has been defined by the outsize influence of Christian conservatives, so much so that the intersection of religion and politics is often treated as the sole province of white evangelicals. And for generations, promises of a rising “religious left” have come and gone without any lasting political imprint.

 

But to look at America’s religious left at this moment is to see something genuinely different. Places of worship are participating in demonstrations for civil rights larger than any protest movement in American history. Democrats like the Rev. Raphael Warnock and Joe Biden—political leaders whose faith isn’t just incidental to their public personas, but is a core component of both their identities and their appeal to voters—are staging important victories. The National Congregations Study, an annual survey of America’s places of worship, found 41 percent of self-identified liberal congregations lobbied or marched about immigration in 2018–19; in 2012, it was only 5 percent. Long locked out of power, a growing religious left is pounding on the door. And it has the potential to remake not only American politics, but the way we think about big questions of fairness, justice and what Americans owe to one another.

 

“Having been a part of the religious left my whole life, yes, it is growing,” says the Rev. Dr. Serene Jones, president of Union Theological Seminary in New York. “Because of the pandemic, people are more open to spirituality in general. And I think that the public eye has been more responsive to seeing the religious left because they need to see them to have any hope at all.”

 

Though the religious left has deep roots in American history—from the abolitionist movement to the establishment of hospitals serving the poor—for much of the past several decades, liberal Christians have been relatively silent about their faith and how it informs their political beliefs, Jones says. But as America grapples with the coronavirus pandemic, yawning economic disparities and the legacy of racism, that’s changing. She points to universal basic income, for example, as “a religious foundational principle” that’s become a lively topic of political debate, and says religious communities are also taking the lead on education and training to combat white supremacy.

 

“Both of those issues are going to continue to escalate in terms of the intensity with which the religious left is facing them,” she says.

 

Of course, for materialists like myself, the key might be to find some common ground in what we can all call "our reality."  


The principles of justice define an appropriate path between dogmatism and intolerance on the one side, and a reductionism which regards religion and morality as mere preferences on the other.   - John Rawls


#289
Raklian

Raklian

    An Immortal In The Making

  • Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 7,517 posts
  • LocationRaleigh, NC

 

 

Think we are living in a simulation.

 

 

On this one point I'd posit that there's no way to know if we are or not.  If you believe humanity will eventually be able to simulate reality then you also must accept that there is a chance we are in that simulated reality.  That said, to us it makes no difference because the simulation IS our reality.

 

 

But it may make a difference if we somehow discover our reality could be turned off any time rather than the Universe gradually dissipating energy over trillions of years. 

 

Maybe you're right it makes no difference since we won't even know our reality has winked out of existence because all of us will simultaneously stop perceiving and thinking.


What are you without the sum of your parts?

#290
Yuli Ban

Yuli Ban

    Born Again Singularitarian

  • Moderators
  • PipPipPipPipPipPipPipPipPipPipPip
  • 22,999 posts
  • LocationNew Orleans, LA

Nvidia's new One-Shot Free-View Neural Talking-Head Synthesis technique

Allows AI-powered video conferencing to use more than 2 times less network bandwidth without sacrificing video quality

For reasons we won't belabor, video conferencing has gained a tremendous user base in the last year. But despite its rise, it's not accessible to many because of the high network bandwidth required to carry both video and speech in real-time. Deep learning techniques (especially GANs) can deliver high-quality video via image compression at lower bit rates. 
But before we dig into this research on a really interesting application of deep learning called talking head synthesis, we recommend checking out the brief video below. It'll help anchor the research as we dig in a bit deeper.


And remember my friend, future events such as these will affect you in the future.


#291
Yuli Ban

Yuli Ban

    Born Again Singularitarian

  • Moderators
  • PipPipPipPipPipPipPipPipPipPipPip
  • 22,999 posts
  • LocationNew Orleans, LA

Robo-writers: the rise and risks of language-generating AI

In June 2020, a new and powerful artificial intelligence (AI) began dazzling technologists in Silicon Valley. Called GPT-3 and created by the research firm OpenAI in San Francisco, California, it was the latest and most powerful in a series of ‘large language models’: AIs that generate fluent streams of text after imbibing billions of words from books, articles and websites. GPT-3 had been trained on around 200 billion words, at an estimated cost of tens of millions of dollars.
The developers who were invited to try out GPT-3 were astonished. “I have to say I’m blown away,” wrote Arram Sabeti, founder of a technology start-up who is based in Silicon Valley. “It’s far more coherent than any AI language system I’ve ever tried. All you have to do is write a prompt and it’ll add text it thinks would plausibly follow. I’ve gotten it to write songs, stories, press releases, guitar tabs, interviews, essays, technical manuals. It’s hilarious and frightening. I feel like I’ve seen the future.”
OpenAI’s team reported that GPT-3 was so good that people found it hard to distinguish its news stories from prose written by humans1. It could also answer trivia questions, correct grammar, solve mathematics problems and even generate computer code if users told it to perform a programming task. Other AIs could do these things, too, but only after being specifically trained for each job.


And remember my friend, future events such as these will affect you in the future.


#292
SeedNotYetSprouted

SeedNotYetSprouted

    Member

  • Members
  • PipPipPipPipPip
  • 202 posts

YOUNG FEMALE TWITTER STAR TURNS OUT TO BE 50-YEAR-OLD MAN USING DEEPFAKES

 

 

You might want to double check that influencer you just followed on Twitter — they might actually be a deepfaked persona.

 
Or at least that was the case with Twitter user @azusagakuyuki, who tricked followers into thinking they were a young female motorbike enthusiast but actually turned out to be a 50-year-old Japanese man named Zonggu using deepfake technology, according to the BBC. 
 
The man behind the Twitter account came clean to his deceptions on a Japanese TV show Monday Late Show, saying he used the FaceApp and other editing tools to create his female persona. Here’s an image from the program of the fake Zonggu and the real Zonggu side by side:
 
3Ck1QQE.jpg
 
 
 

Link



#293
Raklian

Raklian

    An Immortal In The Making

  • Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 7,517 posts
  • LocationRaleigh, NC

Still handsome. LOL


What are you without the sum of your parts?

#294
Yuli Ban

Yuli Ban

    Born Again Singularitarian

  • Moderators
  • PipPipPipPipPipPipPipPipPipPipPip
  • 22,999 posts
  • LocationNew Orleans, LA

Did Myanmar’s military deepfake a minister’s corruption confession?

A detained official and former chief minister of the Yangon region, Phyo Min Thein, appeared on Myawaddy TV, a television network owned and operated by the Myanmar military, on Tuesday. He confessed that he offered bribes to ousted leader Aung San Suu Kyi—or so it seemed.
There was immediate outcry. Journalists and scholars who have met and spoken with Phyo Min Thein noted that the speech in the video did not sound like his real voice. Amateur sleuths and netizens debated whether he was a victim of a deepfake production, where bogus audiovisual content is generated by utilizing deep learning techniques.
“I’ve met U Phyo Min Thein dozens of times and I agree this doesn’t sound like his voice at all,” said historian Thant Myint U on Twitter. Naung Kham, editorial assistant for Channel NewsAsia who has interviewed the former chief minister echoed the sentiment and tweeted, “The voice I hear from the video is different from what I remember.”


 
SYAC: Maybe, but the video quality is too low to have any certainty either way (perhaps deliberately so)

However, if it is true, then yikes. We're in a startling new era of propaganda...


And remember my friend, future events such as these will affect you in the future.


#295
Yuli Ban

Yuli Ban

    Born Again Singularitarian

  • Moderators
  • PipPipPipPipPipPipPipPipPipPipPip
  • 22,999 posts
  • LocationNew Orleans, LA

There’s a New Nirvana Song Out, and It Was Written by Google’s AI

One of the primary capabilities separating human intelligence from artificial intelligence is our ability to be creative—to use nothing but the world around us, our experiences, and our brains to create art. At present, AI needs to be extensively trained on human-made works of art in order to produce new work, so we’ve still got a leg up. That said, neural networks like OpenAI’s GPT-3 and Russian designer Nikolay Ironov have been able to create content indistinguishable from human-made work.
Now there’s another example of AI artistry that’s hard to tell apart from the real thing, and it’s sure to excite 90s alternative rock fans the world over: a brand-new, never-heard-before Nirvana song. Or, more accurately, a song written by a neural network that was trained on Nirvana’s music.
The song is called “Drowned in the Sun,” and it does have a pretty Nirvana-esque ring to it. The neural network that wrote it is Magenta, which was launched by Google in 2016 with the goal of training machines to create art—or as the tool’s website puts it, exploring the role of machine learning as a tool in the creative process. Magenta was built using TensorFlow, Google’s massive open-source software library focused on deep learning applications.


And remember my friend, future events such as these will affect you in the future.


#296
caltrek

caltrek

    Member

  • Members
  • PipPipPipPipPipPipPipPipPipPipPip
  • 12,797 posts

Deepfake Video App Avatarify, Which Processes On=Phone, Plans Digital Watermark for Videos

 

https://techcrunch.c...ark-for-videos/

 

Introduction:

(TechCrunch) Making deepfake videos used to be hard. Now all you need is a smartphone. Avatarify, a startup that allows people to make deepfake videos directly on their phone rather than in the cloud, is soaring up the app charts after being used by celebrities such as Victoria Beckham.

 

However, the problem with many deepfake videos is that there is no digital watermark to determine that the video has been tampered with. So Avatarify says it will soon launch a digital watermark to prevent this from happening.

 

Run out of Moscow but with a U.S. HQ, Avatarify launched in July 2020 and since then has been downloaded millions of times. The founders say that 140 million deepfake videos were created with Avatarify this year alone. There are now 125 million views of videos with the hashtag #avatarify on TikTok. While its competitors include the well-funded Reface, Snapchat, Wombo.ai, Mug Life and Xpression, Avatarify has yet to raise any money beyond an angel round.


The principles of justice define an appropriate path between dogmatism and intolerance on the one side, and a reductionism which regards religion and morality as mere preferences on the other.   - John Rawls


#297
Yuli Ban

Yuli Ban

    Born Again Singularitarian

  • Moderators
  • PipPipPipPipPipPipPipPipPipPipPip
  • 22,999 posts
  • LocationNew Orleans, LA


And remember my friend, future events such as these will affect you in the future.


#298
Yuli Ban

Yuli Ban

    Born Again Singularitarian

  • Moderators
  • PipPipPipPipPipPipPipPipPipPipPip
  • 22,999 posts
  • LocationNew Orleans, LA

VideoGPT: Video Generation using VQ-VAE and Transformers

We present VideoGPT: a conceptually simple architecture for scaling likelihood based generative modeling to natural videos. VideoGPT uses VQ-VAE that learns downsampled discrete latent representations of a raw video by employing 3D convolutions and axial self-attention. A simple GPT-like architecture is then used to autoregressively model the discrete latents using spatio-temporal position encodings. Despite the simplicity in formulation and ease of training, our architecture is able to generate samples competitive with state-of-the-art GAN models for video generation on the BAIR Robot dataset, and generate high fidelity natural images from UCF-101 and Tumbler GIF Dataset (TGIF). We hope our proposed architecture serves as a reproducible reference for a minimalistic implementation of transformer based video generation models. Samples and code are available at this https URL

FINALLY! A breakthrough in true video synthesis. 

We've had to settle for pseudo video synthesis for a couple years now, with existing GAN and autoencoder methods being used on stitched-together animations (you can pull this off in Artbreeder, in fact!) 

But while some of these are indeed novel video synthesis, they really aren't what I've been thinking about when I consider that term. Faces abstractly bleeding into each other or fractals reforming constantly like a DMT trip are certainly interesting to watch, but when I think of "novel video synthesis", I'm thinking of something much more coherent. Take a photograph or a screenshot.

 

Heck, take a screenshot of this page I'm typing on. A novel video synthesis algorithm would be able to "scroll up and down" the page. It doesn't have to know what the hell is on the page besides what's in the screenshot. It just has to remain consistent, creating forum posts that look  at least somewhat legitimate. It's not scraping this forum for that data, either. More that it's "imagining" what the page could be and turning that into a moving sequence of images. 

 

And you can get more complex from there, such as synthesizing an image of a typical street crowd and then animating it, having the people walk and cars move and bikes roll on, maybe even some clouds slowly moving on by. It looks like a video shot on a camera, but this video does not exist.

 

This is the first major step towards it!

I mean, DVD-GANS were interesting, but it's becoming clear that transformers are taking over on this front from GANs.


And remember my friend, future events such as these will affect you in the future.


#299
Yuli Ban

Yuli Ban

    Born Again Singularitarian

  • Moderators
  • PipPipPipPipPipPipPipPipPipPipPip
  • 22,999 posts
  • LocationNew Orleans, LA

More video synthesis news
2021 seems to be what I hoped 2020 to be (if you recall, I was disappointed by the lack of progress in the field of novel video synthesis and text-to-video that year)
 

Microsoft Proposes GODIVA, A Text-To-Video Machine Learning Framework

A collaboration between Microsoft Research Asia and Duke University has produced a machine learning system capable of generating video solely from a text prompt, without the use of Generative Adversarial Networks (GANs).
The project is titled GODIVA (Generating Open-DomaIn Videos from nAtural Descriptions), and builds on some of the approaches used by OpenAI’s DALL-E image synthesis system, revealed earlier this year.


And remember my friend, future events such as these will affect you in the future.






Also tagged with one or more of these keywords: Artificial intelligence, synthetic media, deepfakes, media synthesis, GANs, artificial imagination, image synthesis, natural language generation, automation, deep learning

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users