Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Creating a deepfake took two weeks and cost $552

arstechnica.com

Post reply on HN
Print view View original

41–50 of 119 posts

Re: Creating a deepfake took two weeks and cost $552

#41

Post by mrec »

jml7c5 wrote: post #30
ryanwaggoner wrote: post #8

Earlier quoted context omitted.

I don't really understand this argument at all. We've been in this "post truth" world for a long time already for every form of media other than video. Text quotes, photos, and audio can all be easily faked. If I post a ridiculous quote here and say it's from Obama, you won't believe me. But if the NYTimes does the same thing, it carries a lot more weight. We've been here for a long time already, now those standards…

I don't think it's so much that people will be fooled by fake video, as that people can no longer rely on video as evidence. Up until now, video has been the gold standard of "proof" that an event happened. Quotes, photos, and (to an extent) audio have been fakeable for a long time, and they are mistrusted accordingly. And while CGI has slowly made it harder to spot fake events , videos of speech remained mostly safe…

It'll certainly be weakened, but I don't think it'll necessarily have to be discarded altogether. The need now will be for processes and tools to guarantee the provenance of a video.

The technology to fake paper ballots is trivial, but they still (mostly) work fine for elections, because we've developed strict rules for how they need to be handled in order to retain credibility.

Re: Creating a deepfake took two weeks and cost $552

#42

Post by setr »

adrianN wrote: post #18
ryanwaggoner wrote: post #15

Earlier quoted context omitted.

But why is this the danger now in a way that it hasn't been previously? You could have presented fake documents for hundreds of years now, but it's not like we don't have any credible news organizations as a result.

It gets increasingly cheap to manufacture convincing fakes, but it doesn't get any cheaper to do quality journalism. The risk as I see it is that there is such an overwhelming tsunami of manufactured reality that it becomes impossible to tell truth from fiction.

It was never that expensive to manufacture fake text. Minorly expensive to manufacture fake images.

And we've been hit by tabloids, spam and so on before. None of that is new. Overwhelmed by content is/was/will be a problem that is solved very simply: people limit the distribution lanes they consume, and naturally establish "trust networks" and hierarchies.

Nyt didn't become "trusted" by accident. People aren't as dumb as you seem to think.

Anyways the problem is of distribution, not that a lie can be told. If we can't trust our distribution lanes to actually reflect the institutions we're trusting, then we can't establish our trust networks.

And this is a real problem: our distribution lanes are divorced now from our actual information distrubuters; medium, Facebook, Twitter, etc fuck around with our trust networks, randomly injecting their own bullshit into our feeds and messing around with feed-order based on non-trust metrics (eg money paid) such that they've become fairly unreliable.

And our classical trustable institutions have become less trustworthy, as they flounder about trying to make sense of the "digital age", and have so far done so in a pathetic fashion

Re: Creating a deepfake took two weeks and cost $552

#43

Post by BickNowstrom »

ceejayoz wrote: post #13
ryanwaggoner wrote: post #8

Earlier quoted context omitted.

I don't really understand this argument at all. We've been in this "post truth" world for a long time already for every form of media other than video. Text quotes, photos, and audio can all be easily faked. If I post a ridiculous quote here and say it's from Obama, you won't believe me. But if the NYTimes does the same thing, it carries a lot more weight. We've been here for a long time already, now those standards…

> But if the NYTimes does the same thing, it carries a lot more weight. Sure, and that's part of the danger. Convincing deep fakes will be used to delegitimize mainstream news organizations, as with what happened with Dan Rather - a trusted source presented faked documents.

I think it is the other way around: mainstream news organizations will cry fake news to delegitimize bloggers and amateurs. If everything is easily faked, the only ones to trust with the truth, are the entrenched media companies. Powerful people will be the first to benefit from fake media, when they can claim that the video evidence must be fake ("They photoshopped my arm around her waist!").

Re: Creating a deepfake took two weeks and cost $552

#44

Post by fizixer »

simias wrote: post #4

A few weeks ago my girlfriend made me discover Instagram filters. Obviously they're not in "deep fake" territory but I was still massively impressed by the quality of the face recognition and tracking coupled with some rather advanced 3D reconstructions on top. 15 years ago that would've been science fiction, today it's a free gimmicky feature in an app running on commodity hardware. There are also apps that will smo…

Please don't tell this to mainstream economists though (and their buddies in tech). They don't like to be challenged on the notion that this tech progress is business-as-usual, and that obsolete jobs will be replaced with new kinds of jobs.

Re: Creating a deepfake took two weeks and cost $552

#45

Post by aianus »

ryanwaggoner wrote: post #8
simias wrote: post #4

A few weeks ago my girlfriend made me discover Instagram filters. Obviously they're not in "deep fake" territory but I was still massively impressed by the quality of the face recognition and tracking coupled with some rather advanced 3D reconstructions on top. 15 years ago that would've been science fiction, today it's a free gimmicky feature in an app running on commodity hardware. There are also apps that will smo…

I don't really understand this argument at all. We've been in this "post truth" world for a long time already for every form of media other than video. Text quotes, photos, and audio can all be easily faked. If I post a ridiculous quote here and say it's from Obama, you won't believe me. But if the NYTimes does the same thing, it carries a lot more weight. We've been here for a long time already, now those standards…

> How good will your deep fake have to be in order to fool a deep fake detection AI?

The best generator and the best detector are actually part of the same model! If you create a better detector, you are making the generator better at the same time so you have not accomplished anything.

"GANs are a clever way of training a generative model by framing the problem as a supervised learning problem with two sub-models: the generator model that we train to generate new examples, and the discriminator model that tries to classify examples as either real (from the domain) or fake (generated). The two models are trained together in a zero-sum game, adversarial, until the discriminator model is fooled about half the time, meaning the generator model is generating plausible examples."

https://machinelearningmastery.com/what-are-generative-adver...

Re: Creating a deepfake took two weeks and cost $552

#46

Post by lm28469 »

thepete2 wrote: post #35
simias wrote: post #4

A few weeks ago my girlfriend made me discover Instagram filters. Obviously they're not in "deep fake" territory but I was still massively impressed by the quality of the face recognition and tracking coupled with some rather advanced 3D reconstructions on top. 15 years ago that would've been science fiction, today it's a free gimmicky feature in an app running on commodity hardware. There are also apps that will smo…

What I find most worrying is the 2nd order consequence of not being able to trust pictures anymore.

https://www.history.com/news/josef-stalin-great-purge-photo-...

Photo editing has been used pretty much since the beginning of mainstream photography. You should never "trust" pictures for anything serious if you don't have absolute trust in the source.

Re: Creating a deepfake took two weeks and cost $552

#47

Post by gdubs »

SmallDeadGuy wrote: post #21
xedeon wrote: post #9

That's a terrible Deepfake to be honest. This is a much better example. https://www.youtube.com/watch?v=bPhUhypV27w

More convincing for the face, sure, but last I checked Schwarzenegger had much, much more body mass than in that video. Does anyone know if the technology for faking body shape (and accessories too maybe) is anywhere near as good as for faces yet?

I feel like what makes this one so good is how subtle it is. It took me a moment to go from, “wow, Hader really nailed the facial expressions,” to realizing what was going on.

In this case, the body is being entirely ignored by the autoencoder. The way it works is by first doing a machine vision pass to isolate just faces and process on that region for both sets of source imagery.

Re: Creating a deepfake took two weeks and cost $552

#49

Post by xamuel »

ryanwaggoner wrote: post #8
simias wrote: post #4

A few weeks ago my girlfriend made me discover Instagram filters. Obviously they're not in "deep fake" territory but I was still massively impressed by the quality of the face recognition and tracking coupled with some rather advanced 3D reconstructions on top. 15 years ago that would've been science fiction, today it's a free gimmicky feature in an app running on commodity hardware. There are also apps that will smo…

I don't really understand this argument at all. We've been in this "post truth" world for a long time already for every form of media other than video. Text quotes, photos, and audio can all be easily faked. If I post a ridiculous quote here and say it's from Obama, you won't believe me. But if the NYTimes does the same thing, it carries a lot more weight. We've been here for a long time already, now those standards…

>We've been in this "post truth" world for a long time already for every form of media other than video

The film "Mr. Smith Goes to Washington" came out in 1939. It should be mandatory viewing.

Re: Creating a deepfake took two weeks and cost $552

#50

Post by gdubs »

The central technique used in DeepFakes is fascinating. Initially I assumed they were using Generative Adversarial Networks (GANs). But — and someone can correct if I’m wrong — Deep Fakes use two Autoencoders.

So essentially you’re training a network how to compress an image down to a very tiny representation, and then uncompress it as accurately to the original as possible.

You train two of these: one for the original face, and one for the target face. Then, you compress with the “source” autoencoder, and then uncompress with the “target” autoencoder. And, voila, ‘source‘ face becomes ‘target’ face.

Post reply on HN