＞ SPOOFING

 

Spoofing

Content that poses as a reliable source of information. This category includes content that attempts to manipulate the impressions of recipients of spoofed information by manipulating images and text.
　Since the images and other information that are supposed to be trustworthy and reliable are processed, it is easy to gain trust despite the fact that it is fake news, and it is easy to manipulate the impression.

Japan case

Again, even though the possibility of editing is widely known, the credibility of "images" remains high. Images make people think that they have actually witnessed the scene where the image was taken, and they often consider their opinions based on their unconscious belief that what they see in the image is fact. This is why people are more susceptible to being deceived by fake news when images are manipulated and fakes are mixed with believable images.

＜ A case of manipulating an image to make it look as if it has an inappropriate expression. ＞

In early 2021, an earthquake with a seismic intensity of 6.0 on the Japanese scale struck Fukushima and Miyagi prefectures as they worked to recover from the Great East Japan Earthquake. The government took urgent measures in response and held a press conference. The press conference went without a hitch and seemed to have ended without a hitch. However, after the press conference, a controversial photo of the press conference spread on Twitter, attracting a lot of criticism. The subject of the photo was the then Chief Cabinet Secretary, Katsunobu Kato, who attended and responded to the press conference.
In the photo, Mr. Kato is seen answering questions with a smile next to a ticker tape indicating that he is holding a press conference, despite the urgency and the fact that many people are still suffering. People who saw the photo naturally raised their voices in condemnation, accusing Mr. Kato of being inappropriate. One of the posts with the image attached was retweeted about 2000 times, with comments such as "This is no time to be smiling" and "This is unforgivable".
　The image is a cutout of a televised program of the press conference, and as far as we can tell, there are no unnatural parts found in the photo. It looks as if Mr. Kato actually held the press conference while laughing. However, if you look at the actual video of the press conference, you will not see Mr. Kato smiling during the press conference. It does not mean that the program from which the image was cut out arbitrarily replaced the image of Mr. Kato in a different scene and used it. In creating this image, it was not his intention or the program's editing that altered the image, but the image itself. When the Yomiuri Shimbun questioned the difference between the actual video of the press conference and the disseminated image that is being accused, they asked a professor with digital data analysis skills to analyze it, and found that the eyes and mouth were most likely altered. Mr. Kato had actually held a sincere press conference that was far from inappropriate, and the image that was the cause of the accusations was a "fake" spoofed image.

One of the frightening aspects of this case is that the images were so elaborate that none of the accusers could notice anything wrong with them. According to Professor Uehara, who was asked to evaluate the images Professor Uehara, who appraised the images, said, "It's not easy to distinguish them by elaboration". It is impossible for the general public to distinguish images that even professors, who are far better at appraising images than ordinary people, can only find a few traces of, and the fact that fake images of such sophistication are circulating shows that incidents such as this one can happen at any time. This is also due to the fact that it is no longer difficult to create fake images. In fact, this image was created only 30 minutes after the start of the press conference, and special software and smartphone applications that use AI to alter facial expressions are already in widespread use overseas, making it possible to deceive people with little effort.

Abroad case

As the technology to combat fake news is developing day by day, so are the skills of those who maliciously create fake news. As a result, there is a non-negligible amount of video that has been "edited" to make it look more like the truth, and what should have been believable footage has become unbelievable. And the falsification of "images,"which should be trusted, tends to cause more damage than other fake news items.

< A case of falsification of facts by projecting various celebrities onto another person's video. >

There is something called "Deep Faking". As the name implies, it is a type of fake news, in which artificial intelligence, or AI, can completely replace the faces of people in a video or manipulate the expressions of people in an image at will. In recent years, the quality of this technology has increased dramatically, and it is even possible to create videos that look so natural that they could be mistaken for the person themselves. In addition, the mechanism for creating these videos is so simplified that it is possible to create them without knowledge or ability in video editing. This is why deep fakes are not widely used in Japan, but overseas, deep fake videos are spreading to the extent that they need to be regulated, and the damage is growing in proportion.
　In early 2018, the following video was posted on YouTube in which former President Barack Obama talks about then-President Trump.
video was posted on YouTube in which former President　Barack Obama talks about then-President Trump.→(https://youtu.be/cQ54GDm1eL0)
In this video, Mr. Obama blasted Mr. Trump and even appeared to state that "President Trump is an irredeemable moron." It is hard to believe that Obama would make such a statement since he was the first black U.S. president who made great achievements in desegregation.
However, this seems to be a fact, as the city of Obama in the video does indeed say the words of refuge from his mouth. To tell you the truth, this video was made as an example to show the current deep-faking technology. Created by the American online media company BuzzFeed and actor/director Jordan Peele, the video splits in two as it progresses to the second half, showing President Obama speaking as he did in the first half of the video, and director Peele moving at the exact same time as Obama. All of Obama's lines and movements in this video were made to look as if Obama was actually speaking by reflecting Peele's words and actions on Obama's photo.

　This example was a harmless deep fake, created without malice and intended to draw attention. But, of course, there are many fake news and deep fakes in the world that are made to deceive people with malicious intent, and this video was made because there are many of them, and many of them have the potential to do a lot of damage. In the same way that the AI-enhanced video looked just as good as the real Obama speaking in this video, the maliciously created video also appeared to be true, as if he were actually speaking. Naturally, the technology for countermeasures has evolved to the point where it is on par with the rest of the world, and various countermeasures have been taken, including the use of AI to detect and identify the slightest discomfort, such as a lack of blinking. However, those who create fake news with clear malicious intentions have an advantage over those who take countermeasures in that they can construct countermeasures against countermeasures, and this is why malicious, high-quality deep fakes will never disappear. Therefore, it is important to remember that no matter how true a story may seem, it cannot be considered reliable without cross-referencing multiple sources.

Summary

　

The technology to create fake images and videos is evolving day by day. "It is possible to create fake images" that will fool all the viewers, as in the case of the overseas example, or to edit high quality images in a short time, as in the case of the Japanese example. As technology has improved, so has the ease of creation, and even people without editing skills are now able to create fake images and videos simply by obtaining software from overseas. Therefore, in addition to the two examples introduced here, various fake images and videos have been discovered on the Internet today, and some of them may still remain undiscovered and believed. And that also suggests that there are people who are being accused of innocent crimes because their fake images were believed. Now that we are surrounded by various fake news, what we can do is not to believe the information but to check the authenticity of the information before speaking out.

Related articles

• GRADE
• FABRICATION