Join the movement to end censorship by Big Tech. StopBitBurning.com needs donations and support.
YouTuber uses AI to create "unsettling" deepfakes of South Park characters
By zoeysky // 2024-01-22
Mastodon
    Parler
     Gab
 Copy
 
A YouTuber has used deepfake technology to show what the two-dimensional animated characters in South Park would look like if they were portrayed by real people. Even South Park fans have said that they look strange and downright disturbing. South Park, which premiered in 1997, introduced audiences to the crude world of characters Eric, Kenny, Kyle and Stan, recognizable because of their signature round faces, googly eyes and simplistic, block-colored clothing. The YouTuber who goes by the name @demonflyingfox uploaded a video where the AI program Midjourney was used to transform the foursome, along with other South Park characters, into near-real-life figures. But like other AI-generated images, taking a careful look at them reveals that something is off, from the unnerving blinking to the shifty movements of their heads. After the video was shared on Reddit and the images were also posted on Instagram, many South Park fans commented to offer their opinion on the results. Many did not like the AI-generated video and images. Reddit user Djent_1997 wrote that the "blinking is very unsettling for some reason." Meanwhile, another user named Potheadpig replied that the eerie blinking was because "it’s not natural, you can definitely tell it's not a real person." On Instagram, a user called Austin was at the top of the comment section as he expressed his disdain for AI, saying that the art would have been "cool" if it wasn't made using an AI program. However, others were more positive, with one commenter even saying that they "wish this was real." Deepfakes are videos or images that often feature people who have been digitally altered, whether it's their voice, face or body, to make them look like they are "saying" something else or are someone else entirely. Some examples include a 2019 video showing Tesla cars crashing into a robot at a tech convention, causing havoc. Other deepfake videos include Mark Zuckerberg, where he can be seen allegedly thanking U.S. legislators for their inaction on antitrust issues.

Should you worry about AI and deepfakes?

Although AI and deepfakes don't seem harmful, they can actually be used for fraud and other cybercrimes in the wrong hands. Unfortunately, deep fakes are very easy to create because they have gone mainstream. With many advancements in AI, there are even more accelerated risks to plan for. Companies are especially vulnerable to deepfakes or what is often called disinformation campaigns during public offerings, merger and acquisition transactions and major organizational announcements because these present opportunities where fraudsters can cause havoc and impact a company’s reputation. Social engineering using deepfake methods can carry out targeted phishing attacks (spear phishing) to gain information and data. An attacker can also use AI technology for fraud and to siphon off financial resources. (Related: WARNING: AI-powered DEEPFAKE VOICE SCAMS are now coming for your bank balance.) The Massachusetts Institute of Technology (MIT) built a Detect Fakes website, which is a research project designed to identify techniques to counteract AI-generated misinformation and to advise companies on what to look out for in identifying deep fakes. Follow the tips below to help determine real from deepfake videos.
  1. Look at the face. High-end deepfake edits are almost always facial transformations.
  2. Look at the cheeks and forehead. Does the skin look too smooth or too wrinkly? Is the aging of the skin similar to the aging of their hair and eyes? Deepfakes may be incongruent on some dimensions.
  3. Look at the eyes and eyebrows. Do shadows appear in places that you would expect? Deepfakes may fail to fully represent the natural physics of a scene.
  4. Look at the glasses. Is there any glare? Is there too much glare? Does the angle of the glare on the glasses change when the person moves? Deepfakes are unable to fully represent the natural physics of lighting.
  5. Look at the facial hair or lack thereof. Does the facial hair look real? Deepfakes might add or remove a mustache, sideburns, or beard. However, deepfakes may fail to make facial hair transformations look completely natural.
  6. Look at the facial moles. Does the mole look real?
  7. Look at the blinking. Does the person blink enough or too much?
  8. Look at their lip movements. Some deepfakes are based on lip-syncing. Do the lip movements look natural?
These tips were designed to help guide people looking through deepfakes. While high-quality deepfakes can be hard to spot, with practice, people can build intuition for identifying what is fake and what is real. Visit InformationTechnology.news for more stories about AI and deepfakes. Watch the video below for deepfake videos of Richard Nixon announcing the Apollo 11 disaster. This video is from the SecureLife channel on Brighteon.com.

More related stories:

Former Google CEO presents 6-point plan to combat misinformation – just more ways to MONITOR and CENSOR people. AI anxiety is on the rise as computers get super smart. U.S. government using DEEPFAKES to wage psychological war on the public. Sources include: Metro.co.uk Instagram.com YouTube.com Media.MIT.edu Brighteon.com
Mastodon
    Parler
     Gab
Copy