- Chapo Trap House among leftist channels banned on Twitch for streaming Democratic debate 1 Year Ago
- Meet Ryker, the world’s worst service dog 1 Year Ago
- Far-right blogger claims Trump ordered arrest of Julian Assange 1 Year Ago
- Reddit man wants to tell people he’s been with his girlfriend for one year instead of 6—for an incredibly dumb reason Today 2:18 PM
- John C. Reilly’s son Leo is a TikTok star Today 1:58 PM
- ‘Vanderpump Rules’ recap: A friendship sails Today 1:52 PM
- For celebs, Kobe Bryant tattoos are all the rage Today 1:01 PM
- The internet has discovered Jim and Pam Halpert’s daughter—and she’s on TikTok Today 12:32 PM
- YouPorn launches adult-themed TikTok knock-off Today 12:29 PM
- Clearview AI client list reportedly stolen Today 11:56 AM
- Billie Eilish’s brother Finneas walks back on tone-deaf advice to young creatives Today 11:26 AM
- ‘Caronavirus’ trends after Trump misspells coronavirus Today 11:20 AM
- Conservative writer claims rape victims, trans people are most privileged Today 10:45 AM
- Colin Trevorrow reveals the title of the 3rd ‘Jurassic World’ movie Today 10:28 AM
- You can pre-order Bloomberg’s ‘not a socialist’ hats now if you really want to Today 10:06 AM
British sitcom character Mr. Bean has found himself in yet another deepfake video, this time with Hollywood actors Ben Affleck and Henry Cavill.
The latest altered video, discovered by the Daily Dot in an online deepfakes community, places the face of Mr. Bean star Rowan Atkinson onto Affleck’s and Cavill’s during a 2016 interview for the film Batman v Superman: Dawn of Justice.
Deepfakes are developed using open-source tools that rely on artificial intelligence and machine learning. The technology, which was first introduced to the public by university researchers, has since made its way into the hands of everyday people.
While creating deepfakes does require specific computer knowledge, free online tutorials and tools have placed such capabilities into the hands of anyone willing to devote the time to learn.
Speaking with the Daily Dot, the creator behind the new deepfake, known online as ZeroCool22, explained how the video unfolded.
“I found Mr. Bean’s face to be very ‘unique’ and comical so I wanted to see how it would look on other bodies,” ZeroCool22 said. “In this particular case, I also wanted to see how the neural network would adapt the face of Mr. Bean to two different faces side-by-side.”
Like actor Nicolas Cage, Mr. Bean has become a staple in the deepfakes community. As first reported by the Daily Dot earlier this week, another equally terrifying video saw Mr. Bean’s face placed on that of President Donald Trump’s.
ZeroCool22 says he began creating deepfakes nearly two years ago when the community first emerged in 2017. While plenty videos at the time were jovial in nature, many also centered around placing female celebrities’ faces onto the bodies of porn stars. Those videos, which were of significantly poorer quality than today’s, have mostly disappeared into private forums after widespread bans last year by sites such as Reddit, Pornhub, and Twitter.
Although some videos are more impressive than others in terms of quality, the issue has spurred U.S. lawmakers to raise concerns over the potential malicious use of deepfakes by foreign governments during the 2020 election.
The Bean-Affleck-Cavill mashup is just the latest in a recent string of videos that highlight just how far the technology has come. A high-quality deepfake of actor Steve Buscemi and actress Jennifer Lawrence, first reported on by the Daily Dot last month, appeared to raise the stakes for convincing fake videos.
Numerous lawmakers have proposed legislation aiming to clamp down on deepfakes deemed to be malicious. One bill in particular even called for up to two years in prison for the creator of any deepfake found to incite violence or disrupt an election. Critics state, however, that such legislation is overly broad and could risk violating certain content protected by free speech, such as parody videos.
ZeroCool22 feels that much of the concern is overhyped, but understands that abuses do exist.
“I personally use it just for fun, but it’s like everything, the tech is there, it can be used in a good way or in a bad one,” ZeroCool22 said.
While techniques are being developed to detect and combat deepfakes, ZeroCool22 notes that even more powerful technology is just around the corner. Soon, similar techniques that allow someone to recreate an individual’s voice will likely be as accessible as current deepfake tools.
“Then you will not only be able to fake the appearance, but also the voice,” ZeroCool22 said.
Mikael Thalen is a tech and security reporter based in Seattle, covering social media, data breaches, hackers, and more.