High-tech deception of ‘deepfake’ videos • Hacker News Board

Re: High-tech deception of ‘deepfake’ videos

#31

Post by gnicholas » Mon, Jul 02, 2018, 6:22 PM UTC

mizzack wrote: post #30

gnicholas wrote: post #24

Will FB, Twitter, YouTube, and other social networks create tools to flag deepfakes? It wouldn't solve the problem, but at least it could blunt the spread. It could also raise awareness of the issue, so that if people see a video that seems "too good to be true" on a different platform, they would know that it might not be real.

Inevitably you'll hear cries of censorship with varying degrees of veracity.

I would think that there'd be enough deepfakes on both sides that everyone would recognize that iff there are tools that accurately detect deepfakes, it would be useful to have them enabled on social networks. They don't need to censor the videos — just flag them as having deepfake characteristics.

Re: High-tech deception of ‘deepfake’ videos

#32

Post by noja » Mon, Jul 02, 2018, 6:26 PM UTC

The only solution to this is signed live lifecasting to a blockchain for every person on earth. I'm only half kidding.

Re: High-tech deception of ‘deepfake’ videos

#33

Post by vkou » Mon, Jul 02, 2018, 6:29 PM UTC

isostatic wrote: post #28

SmooL wrote: post #2

I have to agree that this will be inevitable. Fake videos _will_ be created to spread misinformation. I bring you no solutions. Our best programmatic tool for these kinds of video analysis is currently the exact same technology that enables it in the first place. Furthermore, any neural network that can detect fake videos will only be used adversarially to train a better deep fake creator! The only solution I see is…

The last 20 years certain elements have been pushing distrust and hatred for both politicians and journalists. However I'd hope that we move to some agreed standard of signing of videos, so their source can be confirmed by the player. It doesn't matter if that source is the bbc, washington post, alex jones, national enquirer, the onion, or "Bob Joe Billy-Bob the Third", it means you know where the video came from, an…

> A video with a bbc banner showing the president saying something could easily be faked. One that is signed by the BBC will only be fake if the BBC itself is publishing it as a fake, and that comes with a reputational impact.

Nobody cares much about the reputational impact. Fringe tabloid fake-news, and mainstream 'fake'-news gets plenty of viewers.

BBC itself has manufactured video 'evidence' during the Libyan war. [1] Yet, most people think it's an accurate source of news.

[1] Their 'live broadcasts' showed video of people in India waving the Indian tri-color, which the BBC has claimed was rebels in Libya's Green Square waving the old Libyan tri-color. But, eh, who cares about honesty - when you've got a politically sensitive story to tell.

Re: High-tech deception of ‘deepfake’ videos

#34

Post by majos » Mon, Jul 02, 2018, 6:30 PM UTC

avivo wrote: post #21

It's not a totally lost cause — I wrote a piece on what we can do about this: https://www.washingtonpost.com/news/theworldpost/wp/2018/02/... It's a first step, and my thinking has also evolved since—expect more updates in this space soon.

From this article:

> AI researchers and platform technologists need similar in-house or third-party societal impact review boards to help them evaluate the potential unintended consequences of their work.

As a machine learning researcher myself (albeit one who is probably unlikely to run afoul of one of these boards), this suggestion for research gives me pause. In the context of deepfakes, are you suggesting that it would have been more prudent to stop the research that made it possible before it had a chance to do so? This seems like a good way to ensure that less scrupulous parties harness destructive technologies before more scrupulous parties (the kind who'd adopt review boards).

I imagine this debate to some extent has already played out with nuclear weapons and bioweapons. I'm much less familiar with these two spheres, but as I recall research continued in both.

Re: High-tech deception of ‘deepfake’ videos

#35

Post by JumpCrisscross » Mon, Jul 02, 2018, 6:30 PM UTC

SmooL wrote: post #2

I have to agree that this will be inevitable. Fake videos _will_ be created to spread misinformation. I bring you no solutions. Our best programmatic tool for these kinds of video analysis is currently the exact same technology that enables it in the first place. Furthermore, any neural network that can detect fake videos will only be used adversarially to train a better deep fake creator! The only solution I see is…

> the amount of cooperation we'd need between camera makers and video distribution platforms

Apple could single-handedly launch a "verified video" system.

Re: High-tech deception of ‘deepfake’ videos

#36

Post by mizzack » Mon, Jul 02, 2018, 6:30 PM UTC

gnicholas wrote: post #31

mizzack wrote: post #30

Earlier quoted context omitted.

Inevitably you'll hear cries of censorship with varying degrees of veracity.

I would think that there'd be enough deepfakes on both sides that everyone would recognize that iff there are tools that accurately detect deepfakes, it would be useful to have them enabled on social networks. They don't need to censor the videos — just flag them as having deepfake characteristics.

Like when Facebook (briefly) tagged articles from certain sources as being disreputable and it led to an _increase_ in clicks to those articles?

People will believe what they want to believe. We're also seeing that they don't like mega corps acting as arbiters of truth. So, I'm not sure what, if any, effect this moderation/curation effort would have.

Re: High-tech deception of ‘deepfake’ videos

#37

Post by mjw_byrne » Mon, Jul 02, 2018, 6:32 PM UTC

Deepfakes are surely no more a "problem" than photoshopped still images?

Everyone knows that a still image can be altered to (convincingly) depict something that never happened. So, a still image is not considered good evidence of anything unless it has a verifiable provenance, chain of custody etc. Presumably that wasn't always the case, but now that photo editing tech is universally accessible, we all know we can't trust a still image unless its authenticity can be corroborated somehow.

The deepfake tech merely puts video into the same category of (un)trustworthiness. The potential for abuse depends on people's trust that video is always authentic. So the "solution" is simply to make it as widely-known as possible that video is now easy to fake.

Re: High-tech deception of ‘deepfake’ videos

#38

Post by avivo » Mon, Jul 02, 2018, 6:33 PM UTC

SmooL wrote: post #2

I have to agree that this will be inevitable. Fake videos _will_ be created to spread misinformation. I bring you no solutions. Our best programmatic tool for these kinds of video analysis is currently the exact same technology that enables it in the first place. Furthermore, any neural network that can detect fake videos will only be used adversarially to train a better deep fake creator! The only solution I see is…

Unfortunately, even if we can pull it off "cryptographically signing a video" doesn't "solve" the problem. How would that verification be presented to a user? How/why would it influence their beliefs/actions?

(As mentioned elsewhere in this thread, I go into a bit more detail here: https://www.washingtonpost.com/news/theworldpost/wp/2018/02/... )

Will try to write a HN friendly tldr/update soon.

Re: High-tech deception of ‘deepfake’ videos

#39

Post by sp332 » Mon, Jul 02, 2018, 6:34 PM UTC

creaghpatr wrote: post #6

SmooL wrote: post #2

I have to agree that this will be inevitable. Fake videos _will_ be created to spread misinformation. I bring you no solutions. Our best programmatic tool for these kinds of video analysis is currently the exact same technology that enables it in the first place. Furthermore, any neural network that can detect fake videos will only be used adversarially to train a better deep fake creator! The only solution I see is…

The only thing that can be done is use contextual evidence to disprove the possibility of a video being real but that puts a disproportionate on the victim of a fake video. Will the government be in the business of commenting on the veracity of whatever video(s) they anticipate coming? And if they say it's fake (or real), will the public believe them?

It will go back to being like a newspaper, where you can fake-quote anybody saying anything. Only established sources with some reputation on the line will be trustable. A random citizen won't be able to whip out a cell phone and get evidence that will be taken seriously. Credibility will be more centralized again.

Re: High-tech deception of ‘deepfake’ videos

#40

Post by sp332 » Mon, Jul 02, 2018, 6:36 PM UTC

zzzcpan wrote: post #15

bytematic wrote: post #4

Earlier quoted context omitted.

Can we develop something to put in videos that are real? Maybe something built into the software or maybe even some icon you have to hold up behind someone like the president to at least make deepfakes difficult/impossible? Just spit-balling as I also see no point in video analysis

We already have developed something. Only thing missing now is cultural shift in the media to apply a little bit of effort into signing videos.

That might be solving the wrong problem. You could believe that a video is from a particular journalist, but you might not trust that the journalist didn't tamper with the video before publishing it.