'Deep fakes': How to know what's true in the fake-Obama video era

mirimir · Jun 6, 2019

“This technology is really about better storytelling,” Fried said.
Click to expand...

We can all agree on that

guest · Jun 10, 2019

AI deepfakes are now as simple as typing whatever you want your subject to say
A scarily simple way to create fake videos and misinformation
June 10, 2019
https://www.theverge.com/2019/6/10/18659432/deepfake-ai-fakes-tech-edit-video-by-typing-new-words

In the latest example of deepfake technology, researchers have shown off new software that uses machine learning to let users edit the text transcript of a video to add, delete, or change the words coming right out of somebody’s mouth.
The work was done by scientists from Stanford University [...] This work is just at the research stage right now and isn’t available as consumer software, but it probably won’t be long until similar services go public.

In a blog post accompanying the paper, they write: “Although methods for image and video manipulation are as old as the media themselves, the risks of abuse are heightened when applied to a mode of communication that is sometimes considered to be authoritative evidence of thoughts and intents. [...]"
The researchers note that technology like this has many beneficial uses, too. It would be of great help to the film and TV industries, allowing them to fix misspoken lines without rerecording footage, and create seamless dubs of actors speaking different languages.
Click to expand...

mirimir · Jun 10, 2019

Yes, that is a very disturbing development. But as the article and paper argue, maybe the benefits outweigh the risks.

Still, the bottom line is that you can't rely on video as factual evidence.

hawki · Jun 11, 2019

"This Deepfake of Mark Zuckerberg Tests Facebook’s Fake Video Policies

A fake video of Mark Zuckerberg giving a sinister speech about the power of Facebook has been posted to Instagram. The company previously said it would not remove this type of video...

The video, created by artists Bill Posters and Daniel Howe in partnership with advertising company Canny, shows Mark Zuckerberg sitting at a desk, seemingly giving a sinister speech about Facebook's power...

'Imagine this for a second: One man, with total control of billions of people's stolen data, all their secrets, their lives, their futures,' Zuckerberg's likeness says, in the video. 'I owe it all to Spectre. Spectre showed me that whoever controls the data, controls the future.'..."

Story and Video here:

https://www.vice.com/en_us/article/ywyxex/deepfake-of-mark-zuckerberg-facebook-fake-video-policy

guest · Jun 12, 2019

New tool debunks deepfakes of Trump and other world leaders
[...] researchers detected all kinds of video forgeries
June 12, 2019
https://www.cnet.com/news/deepfake-detection-tool-debunks-fakes-of-trump-and-other-world-leaders/

Deepfakes of world leaders may be easier to debunk using a new detection method, according to an academic paper Wednesday.
That "soft biometric model" helped detect a range of deepfakes, the kind of manipulated videos powered by artificial intelligence that have sprung up lately featuring Mark Zuckerberg and others.
(The rest of us are still out of luck if anybody makes a deepfake of us, though.)

The three main types of deepfakes -- referred to here as face swaps, lip syncs and puppet-master fakes -- all rely on rewriting how a victim's entire face or mouth moves. But all the subtleties of that victim's pseudo-fingerprint are too hard to replicate even by the best impersonator. So the researchers are able to apply this fingerprint they developed to distinguish between real and fake videos.

The latest paper -- which was based on research funded by Google, Microsoft and DARPA -- is titled "Protecting World Leaders Against Deep Fakes" and published by researchers at the University of California, Berkeley, and the University of Southern California.
Click to expand...

guest · Jun 12, 2019

hawki said: ↑

"This Deepfake of Mark Zuckerberg Tests Facebook’s Fake Video Policies
Click to expand...

Facebook won't yank Mark Zuckerberg deepfake video from Instagram
June 12, 2019
https://www.theinquirer.net/inquirer/news/3077298/mark-zuckerberg-deepfake-instagram

The social network Facebook has decided to leave a 'deepfake' video of bossman Mark Zuckerberg on Instagram.

"We will treat this content the same way we treat all misinformation on Instagram," said a spokesman. "If third-party fact-checkers mark it as false, we will filter it from Instagram's recommendation surfaces like Explore and hashtag pages."
Taking that 'all things are equal' approach is arguably a noble one, as Facebook isn't pushing Instagram fact-checkers to drop their work in order to get rid of a troublesome post just because the big boss is in it.
Click to expand...

hawki · Jun 13, 2019

"House Intelligence Committee To Hold Open Hearing on Deepfakes and AI

The National Security Challenge of Artificial Intelligence, Manipulated Media, and “Deepfakes”

Washington, DC – On Thursday, June 13, 2019 at 9:00 am, the House Permanent Select Committee on Intelligence will convene an open hearing on the national security challenges of artificial intelligence (AI), manipulated media, and “deepfake” technology. This is the first House hearing devoted specifically to examining deepfakes and other types of AI-generated synthetic data..."

https://intelligence.house.gov/news/documentsingle.aspx?DocumentID=657

hawki · Jun 13, 2019

"Spy used AI-generated face to connect with targets [on Linkden]

LONDON (AP) — Katie Jones sure seemed plugged into Washington’s political scene...

But Katie Jones doesn’t exist, The Associated Press has determined. Instead, the persona was part of a vast army of phantom profiles lurking on the professional networking site LinkedIn...

Experts who reviewed the Jones profile’s LinkedIn activity say it’s typical of espionage efforts on the professional networking site, whose role as a global Rolodex has made it a powerful magnet for spies...

...[F]oreign spies routinely use fake social media profiles to home in on American targets — and accused China in particular of waging “mass scale” spying on LinkedIn...

British , French and German officials have all issued warnings over the past few years detailing how thousands of people had been contacted by foreign spies over LinkedIn...

The Katie Jones profile was modest in scale, with 52 connections. But those connections had enough influence that they imbued the profile with credibility to some who accepted Jones’ invites...

Klingemann and other experts said the photo ...appeared to have been created using a family of dueling computer programs called generative adversarial networks, or GANs, that can create realistic-looking faces of entirely imaginary people. GANs, sometimes described as a form of artificial intelligence, have been the cause of increasing concern for policymakers already struggling to get a handle on digital disinformation..."

https://www.apnews.com/bc2f19097a4c4fffaa00de6770b8a60d

How GANs work:

"Style-based GANs – Generating and Tuning Realistic Artificial Faces..."

https://www.lyrn.ai/2018/12/26/a-st...itecture-for-generative-adversarial-networks/

hawki · Jun 13, 2019

hawki said: ↑

"House Intelligence Committee To Hold Open Hearing on Deepfakes and AI

The National Security Challenge of Artificial Intelligence, Manipulated Media, and “Deepfakes”

Washington, DC – On Thursday, June 13, 2019 at 9:00 am, the House Permanent Select Committee on Intelligence will convene an open hearing on the national security challenges of artificial intelligence (AI), manipulated media, and “deepfake” technology. This is the first House hearing devoted specifically to examining deepfakes and other types of AI-generated synthetic data..."

https://intelligence.house.gov/news/documentsingle.aspx?DocumentID=657
Click to expand...

FYI:

Live Broadcast - 9:00 AM

https://www.c-span.org/video/?46167...l-security-risks-artificial-intelligence&live

Recorded video will be available for 24+ hours at

https://www.c-span.org/

hawki · Jun 13, 2019

"Top AI researchers race to detect ‘deepfake’ videos: ‘We are outgunned’

Top artificial-intelligence researchers across the country are racing to defuse an extraordinary political weapon: computer-generated fake videos that could undermine candidates and mislead voters during the 2020 presidential campaign.

And they have a message: We’re not ready...

The researchers have designed automatic systems that can analyze videos for the telltale indicators of a fake, assessing light, shadows, blinking patterns — and, in one potentially groundbreaking method, even how a candidate’s real-world facial movements — such as the angle they tilt their head when they smile — relate to one another.

But for all that progress, the researchers say they remain vastly overwhelmed by a technology they fear could herald a damaging new wave of disinformation campaigns, much in the same way fake news stories and deceptive Facebook groups were deployed to influence public opinion during the 2016 election..."

https://www.washingtonpost.com/tech...deos-we-are-outgunned/?utm_term=.ee644c06b32a

guest · Jun 14, 2019

Adobe Shows Off First Research for Tools to Detect Manipulated Photos
June 14, 2019
https://gizmodo.com/adobe-shows-off-first-research-for-tools-to-detect-mani-1835516127

[...] in an attempt to counter the fake news epidemic, Adobe has also started developing tools that can both detect when an image has been manipulated, and reverse the changes to reveal the original.
This new detection tool has also been designed with a hyper-specific task: to detect when a face has been warped and subtly manipulated using Photoshop’s own Face Aware Liquify tool [...]

The neural network, by comparison, was able to find the manipulated image 99 percent of the time. Furthermore, the detection tool was able to tell where and how a face had been warped, and even undo those changes and revert the photos to their original state.
Those are promising results, but this neural network has a long way to go before it could be unleashed upon the billions of photos floating around the internet right now.
Click to expand...

guest · Jun 23, 2019

Long read
The rise of the deepfake and the threat to democracy
June 22, 2019
https://www.theguardian.com/technol...e-of-the-deepfake-and-the-threat-to-democracy

Recent electoral upsets have demonstrated the unprecedented power of political entities to microtarget individuals with news and content that confirms their biases. The incentive to use deepfakes to injure political opponents is great.
Some question the scale of this threat. Russell Brandom, policy editor at the Verge, the US tech news site, argued recently that deepfake propaganda is “a crisis that doesn’t exist” [...] But many experts disagree.

The most sophisticated deepfakes require advanced machine-learning skills and their development is computationally intensive and expensive. One expert estimates the cost to be about £1,000 a day. For an amateur creating fake celebrity pornography, this is a major barrier to entry. But for a government or a well-funded political organisation, the cost is insignificant – and falling every month.

Media literacy can only go so far; humans often believe first, then look for things that support those beliefs.
“We have to inoculate the public before deepfakes affect elections,” she says.
Click to expand...

guest · Jun 23, 2019

Samsung’s deepfake technology is getting scary
June 23, 2019
https://bgr.com/2019/06/23/deepfake-ai-samsung-rasputin-beyonce/

Samsung researchers have been particularly adept at this, with the latest example being the partnership between Samsung’s AI research center in the UK working with Imperial College London to create an AI that animates and syncs up an audio clip with facial movements derived from little more than a photograph of someone.

All of the technical work that went into the process is explained here, but for a look at their work in action, behold. This is a clip that uses a grainy black-and-white photo of the dead Russian mystic and self-proclaimed monk Grigori Rasputin to add facial movements that make it look like he’s singing Beyonce’s song Halo.

“The videos generated using this model do not only produce lip movements that are synchronized with the audio but also exhibit characteristic facial expressions such as blinks, brow raises, etc.,”

The Rasputin clip above isn’t realistic at all, of course, but it gives you a sense of how developed this technology is becoming.
Click to expand...

guest · Jun 27, 2019

mood said: ↑

Facebook won't yank Mark Zuckerberg deepfake video from Instagram
June 12, 2019
https://www.theinquirer.net/inquirer/news/3077298/mark-zuckerberg-deepfake-instagram
Click to expand...

Zuckerberg says company ‘evaluating’ deepfake video policy
June 27, 2019
https://apnews.com/506bfacda55540e1921a2dd101926fdf

Facebook CEO Mark Zuckerberg says the company is evaluating how it should handle “deepfake” videos created with artificial intelligence and high-tech tools to yield false but realistic clips.

In an interview at the Aspen Ideas Festival in Colorado on Wednesday, Zuckerberg said it might make sense to treat such videos differently from other misinformation such as false news.
But Zuckerberg says it’s worth asking whether deepfakes are a “completely different category” from regular false statements. He says developing a policy on these videos is “really important” as AI technology grows more sophisticated.

Company executives have said in the past that it makes sense to look at them under the broader umbrella of false or misleading information. But Zuckerberg is signaling that this view might be changing, leaving open the possibility that Facebook might ban deepfakes altogether.
Click to expand...

Victek · Jun 27, 2019

mood said: ↑

Long read
The rise of the deepfake and the threat to democracy
June 22, 2019
https://www.theguardian.com/technol...e-of-the-deepfake-and-the-threat-to-democracy
Click to expand...

Excellent article.

"Whatever the creators of deepfakes and the software that builds them may say, the loss of citizen confidence in the trustworthiness of information is destructive for democracy."

hawki · Jun 27, 2019

"New AI deepfake app creates nude images of women in seconds

The resulting fakes could be used to shame, harass, and intimidate their targets

A new AI-powered software tool makes it easy for anyone to generate realistic nude images of women simply by feeding the program a picture of the intended target wearing clothes.

The app is called DeepNude and it’s the latest example of AI-generated deepfakes being used to create compromising images of unsuspecting women..."

https://www.theverge.com/2019/6/27/...app-women-deepnude-non-consensual-pornography

https://www.vice.com/en_us/article/kzm59x/deepnude-app-creates-fake-nudes-of-any-woman

guest · Jun 28, 2019

Creator of DeepNude, App That Undresses Photos of Women, Takes It Offline
June 27, 2019
https://www.vice.com/en_us/article/...at-undresses-photos-of-women-takes-it-offline

The creator of DeepNude, an app that used a machine learning algorithm to "undress" images of clothed women announced Thursday that he's killing the software, after viral backlash for the way it objectifies women.
Following Motherboard's story, the server for the application, which was available for Linux and Windows, crashed

By Thursday afternoon, the DeepNude twitter account announced that the app was dead: No other versions will be released and no one else would be granted to use the app.
"We thought we were selling a few sales every month in a controlled manner... We never thought it would become viral and we would not be able to control traffic."
Click to expand...

The Verge: Controversial deepfake app DeepNude shuts down hours after being exposed

Circuit · Jun 29, 2019

https://www.france24.com/en/20190628-deepnude-app-undress-women-shut-down-after-furor

Dam, never got a chance to use.

reasonablePrivacy · Jun 30, 2019

Circuit said: ↑

https://www.france24.com/en/20190628-deepnude-app-undress-women-shut-down-after-furor

Dam, never got a chance to use.
Click to expand...

Well, it turns out you must use your own imagination to do what that app do.

mirimir · Jun 30, 2019

Given common clothing styles, not much imagination is required.

Minimalist · Jul 1, 2019

Deepfake revenge porn distribution now a crime in Virginia

As of today, Virginia is one of the first states in the country to impose criminal penalties on the distribution of non-consensual "deepfake" images and video.
Click to expand...

https://arstechnica.com/tech-policy...ge-porn-distribution-now-a-crime-in-virginia/

guest · Jul 9, 2019

Github is banning copies of ‘deepfakes’ porn app DeepNude
July 9, 2019
https://www.theverge.com/2019/7/9/20687902/github-bans-deepnude-deepfakes-ai-nudity-app-copies

GitHub is banning code from DeepNude, the app that used AI to create fake nude pictures of women. Motherboard, which first reported on DeepNude last month, confirmed that the Microsoft-owned software development platform won’t allow DeepNude projects.
GitHub told Motherboard that the code violated its rules against “sexually obscene content,” and it’s removed multiple repositories, including one that was officially run by DeepNude’s creator.

None of this can stop copies of DeepNude from appearing online — but GitHub’s decision could make the app harder to find and its algorithm harder to tinker with.
Click to expand...

hawki · Jul 12, 2019

"Fake DeepNude Downloads Gives You Malware Instead of Nudes

Cybercriminals are using the notoriety of the DeepNude app to distribute info-stealing malware in campaigns over YouTube that promise a cracked premium version of the program for Windows, Android, and iOS..."

https://www.bleepingcomputer.com/ne...downloads-gives-you-malware-instead-of-nudes/

hawki · Jul 19, 2019

"Criminals are using deepfakes to impersonate CEOs...

...[D]eepfakes are now moving past the porn realm and into the criminal world where bad actors are using the tech to impersonate CEOs,... However, for now, it appears criminals are using deepfake audio instead of video to pull off scams:

Symantec, a major cybersecurity company, says it has seen three successful audio attacks on private companies. In each, a company’s “CEO” called a senior financial officer to request an urgent money transfer.

Scammers were mimicking the CEOs’ voices with an AI program that had been trained on hours of their speech—culled from earnings calls, YouTube videos, TED talks, and the like.

Millions of dollars were stolen from each company, whose names were not revealed...

...[T]he tech could also be used for sabotage. What if one rival–or even a nation-state–wanted to sink Apple’s stock price? A well-timed deepfake audio clip that purports to show Tim Cook having a private conversation with someone about iPhone sales tanking could do just that–wiping billions off the stock market in seconds..."

https://www.fastcompany.com/90379001/criminals-are-using-deepfakes-to-impersonate-ceos

guest · Jul 20, 2019

AI vs. AI: This Neural Network Can Spot DeepFake Photos
July 20, 2019
https://fossbytes.com/neural-network-spot-deepfake-photos/

We’re entering a new era where DeepFakes will be common and if means are not invented to control them, it will be a menace. On similar lines, researchers at the University of California have developed a deep neural network that can spot deepfake photos.

The neural network is developed to spot patterns in raw data and is modeled similar to a human brain.
While most of the time, it is not possible to detect morphed photos by the naked eye, but a computer that could examine the photos pixel-by-pixel can use this fact to spot the deepfakes.
After the neural network was shown images outside of the dataset previously fed to it, the network was able to spot deepfakes “most of the time.”

The results were satisfactory but the neural network currently works only for photos. Researchers are figuring out a way to apply it to deepfake videos as well.
However, we cannot say that this is a fix for deepfakes as the neural network is not 100% accurate.
Click to expand...

Log in or Sign up

'Deep fakes': How to know what's true in the fake-Obama video era

mirimir Registered Member

guest Guest

mirimir Registered Member

hawki Registered Member

guest Guest

guest Guest

hawki Registered Member

hawki Registered Member

hawki Registered Member

hawki Registered Member

guest Guest

guest Guest

guest Guest

guest Guest

Victek Registered Member

hawki Registered Member

guest Guest

Circuit Registered Member

reasonablePrivacy Registered Member

mirimir Registered Member

Minimalist Registered Member

guest Guest

hawki Registered Member

hawki Registered Member

guest Guest

Log in or Sign up

'Deep fakes': How to know what's true in the fake-Obama video era

mirimir Registered Member

guest Guest

mirimir Registered Member

hawki Registered Member

guest Guest

guest Guest

hawki Registered Member

hawki Registered Member

hawki Registered Member

hawki Registered Member

guest Guest

guest Guest

guest Guest

guest Guest

Victek Registered Member

hawki Registered Member

guest Guest

Circuit Registered Member

reasonablePrivacy Registered Member

mirimir Registered Member

Minimalist Registered Member

guest Guest

hawki Registered Member

hawki Registered Member

guest Guest

Useful Searches