With deepfakes getting more and more sophisticated, it’s harder than ever to tell the difference between what’s real and what isn’t. Today we will tell you about some striking signs that will help you detect deep fake videos. With these tools at your disposal, we hope that you will not be an unwitting victim of fake news and conspiracy theories. We will also give you an insight into deepfake technology, how it works and how it is used by propagandists and conspiracy theorists to spread disinformation and make netizens believe the lies.
Detect Deepfake Videos to Prevent Disinformation and Conspiracy Theories
Currently, deep scams are largely used by amateur hobbyists to change faces of celebrities in adult videos, and by unsavory political elements to spread fake news. But experts are worried about even more dangerous uses of technology in the not-too-distant future. So here we list some ways to help you detect deep fake videos to avoid falling victim to malicious disinformation.
What are Deepfake Videos?
Deep scams, a combination of the terms ‘deep learning’ and ‘fake’, refer to manipulated media that uses artificial intelligence or deep learning techniques to create audio or videos that distort reality. The technology uses artificial neural networks to create hyper-realistic videos that show people saying or doing something they wouldn’t do in real life. The most violent examples include videos formatted with the help of machine learning to get the words into politicians’ mouths. This is used to create confusion about their policies and influence elections.
Another toxic use of deepfakes is replacing celebrities’ faces in adult videos with massive breaches of privacy and dignity. Deepfakes have become a huge problem in the last few years and are getting worse with the availability of faster hardware and more accurate software. The technology has recently gained notoriety for its widespread use in pornographic videos, fake news, and elaborate hoaxes.
But not all uses of deepfakes are shady, as the video below by MIT scholar Alexander Amini to make his students laugh. It seems that former US President Barack Obama is inviting students to attend Amini’s lecture on deep learning at the university. However, as the researcher made clear on his YouTube channel, it is a deep fraud from start to finish.
How Deepfakes Work
Deepfakes relies on an artificial neural network called an ‘autocoder’, which is used to learn efficient data encodings in an unsupervised manner. It is often used for face recognition and also to find the semantic meanings of words. In the case of deep fake videos, the technology first uses an encoder to train a neural network on hours of real video footage of the target person. . Next, a decoder creates a new image using key information about facial features and body posture. This helps the algorithm superimpose the target’s facial and physical features on the subject in the original video.
A well-known technology in this context is a special class of deep learning algorithms called the generative adversarial network. GAN is often added to the decoder for more accurate results. A GAN trains the decoder and a discriminator so that the former generates new images from the source material, while the latter determines whether the newly created image matches the actual image. This causes the renderer to render images that mimic reality extremely well, because the competing algorithm catches any flaws.
This makes it extremely difficult to fight deepfakes as they are constantly evolving. When any error occurs, it can be corrected automatically through machine learning. Since GAN doesn’t require any human input, it has become the ultimate choice for most deep dishonest content creators. However, the technology is complex and requires much more time and data to create realistic compositions. Also, while GANs are fine for synthesizing images, they have a hard time maintaining temporal consistency, meaning they need human intervention to maintain the alignment of images from one frame to the next.
What Are Shallow Fake Videos?
A shallow fake video is a falsified version of an existing real video, created to reflect a distorted reality. This often includes selective editing, manipulating the pace of people speaking or speaking, and even changing the tonality to make it seem like someone is angry, drunk, or making fun of a serious problem when none of this is true.
One recent notable case involves the notorious fine-tuned video of Nancy Pelosi, the speaker of the US House of Representatives, whose shallow-fake video slows her speech, making her voice intoxicating. These videos differ from deep scammers as they are real videos manipulated using traditional video editing tools instead of AI algorithms.
Deepfakes: History and Applications
Photo manipulation techniques were first invented in the late 19th century. Before the explosion in artificial intelligence and machine learning technologies, technology has evolved steadily throughout the twentieth century, making it a huge issue for netizens worldwide. AI-infused video manipulation techniques have been widely studied by researchers since the 1990s, and many of the methods have since been adopted by filmmakers around the world.
One of the best-known examples of deepfakes used in the mainstream entertainment industry was the resurrection of the late actor Paul Walker. Fast and Furious 7 However, while it took dozens of experts several weeks to recreate Walker believably, it now takes only hours (or sometimes less) for most hobbyists with little coding knowledge to create new deep fake videos using new techniques and algorithms. The phenomenon first entered the public consciousness in 2017 when a Redditor used deepfakes to create and post fake porn videos of celebrities.
Dangers of Deepfake Videos
Deepfake videos are a danger to unsuspecting users, who can be bombarded with footage of a supposedly destabilizing event such as a war or terrorist attack that never took place. It can cause resentment and discontent in the society, leading to an increase in politically motivated attacks based on people’s racial, religious and ethnic identities. Technology can also be used to spread FUDs (fear, uncertainty, and suspicion) about natural disasters, causing widespread panic. Experts also speculate that if left unchecked, such videos could lead to deep political crises and even disrupt international relations.
Another important problem that has already reached pandemic proportions is the threat to unsuspecting women. Deep fake adult videos, often referred to as non-consensual pornography, were reportedly responsible more than 90% While it started with disfigured videos depicting celebrities like Gal Gadot and Alexandra Daddario, it has since expanded to target ordinary women as part of fake revenge porn campaigns.
How to Detect Deepfake Videos?
Detecting deep fake videos is a task that even experts often find difficult without the right tools. However, researchers at the Massachusetts Institute of Technology (MIT) have made a few suggestions that may help ordinary people understand the difference between real videos and deep fakes. According to them, one has to be very careful with the face when trying to check whether a video of a human subject is real or fake. This is because high-level deep fake manipulations are almost always face transformations.
This The areas of the face that need attention the most are the cheeks and forehead. Does the skin look too smooth or too wrinkled? Is the age of the skin similar to the age of your hair and eyes? “DeepFakes are often incompatible with some dimensions”, say the researchers. Similarly, eyes and eyebrows can be telltale signs for seasoned deep fake watchers. Because, according to researchers, Shadows in deepfake videos don’t always appear where you expect them to. “DeepFakes often cannot fully represent the natural physics of a scene”they say.
Another feature that is a dead giveaway is the beard. Deepfakes can add or remove mustaches, sideburns or beards, but they often cannot make facial hair transformations look completely natural. It’s the same with facial moles, which often don’t look natural enough in deepfakes. The size and color of the lips can also give a clue about the validity of a video. Blink rate and speed can say a lot about whether a video is real or fake. Unnaturally frequent or infrequent flashing can indicate deep falsification of a video.
According to MIT researchers, detecting high-quality deep frauds is not easy, but “With practice, people can develop intuition to determine what is fake and what is real”. The researchers also created a full-fledged web page where people can upload videos and try to guess if they are real or fake. You can try your deepfake detection skills at MITs Fake Detection Web site.
Deepfakes: Prevention and Legislative Action
Various countries around the world are trying to address the clear and present danger posed by deep frauds infused with AI. While China banned deepfake videos in 2019, the US state of California passed a similar law earlier that same year to outlaw political deepfakes, banning the creation or distribution of tampered videos, images or sounds of politicians within 60 days of the election. . Since then, other US states, including Texas and Virginia, have also criminalized deep fake porn. In December 2019, President Trump signed the nation’s first federal law to combat deepfakes as part of the 2020 National Defense Authorization Act.
By the way, there is no specific law on deep fake media in India. In fact, the laws regarding AI algorithms are sketchy at best. One of the most notable uses of deepfake in the country was seen during the 2020 Delhi Elections, when BJP’s BT Cell released an official campaign video featuring prime minister candidate Manoj Tiwari addressed to the voters in Hindi, Haryanvi and English. . The problem is that only the Hindi video was real while the other two were deep fake clips produced using the original video to reach a wider audience.
Prevent the Spread of Disinformation by Detecting Deepfake Videos
The guardians of multi-million dollar Hollywood productions and state-sponsored agencies and organizations, deepfakes have become increasingly democratic lately. This allows normal netizens to create deepfake vaults using deepfake apps and websites. With deepfakes increasing astronomically in recent years, being able to detect them is more important than ever.
We hope the information here has helped give you a more holistic view of the technology, the threats it poses, and the signs you should look out for to better detect deep fake videos in the future. But have you ever been a victim of deep fraud from disgruntled political action groups or conspiracy theorists? Let us know in the comments below.