Why did Bill Gates cross the road?
We don't know, but we can assure you that the Microsoft (MSFT) founder didn't have breast implants.
'In Other Fact Check News'
A digitally altered video has been popping up on social media that depicts the billionaire with an enlarged chest as he crosses a street in New York.
The video clip, which recently resurfaced after appearing online last year, sparked questions about Gates "transitioning" or receiving breast implants.
Reuters Fact Check declared the video altered, by comparing the clip to unedited footage that was published on YouTube by entertainment site Gossip Bae.
The April 25 announcement was greeted with some serious sarcasm when it was posted on Twitter.
"Holy sh*t, did u guys know that game of thrones wasn't real too??" one commenter said, including a CGI-free clip from the popular HBO series.
"In other fact check news, the dinosaurs in Jurassic Park were not real and were in fact a combination of CGI and animatronic creations," one person tweeted.
'An Inflection Point'
Other commenters brought up the name of Tesla (TSLA) CEO Elon Musk, who is acquiring Twitter (TWTR) for $44 billion and taking the company private.
"Enjoy your last days of relevance on this app," one person said. "For you will be a relic of social media once Elon buys out Twitter."
"I pray that when @elonmusk owns Twitter we have to see less of these ridiculous 'fact checks' of nonsensical stuff," another person said, adding. "'Fact-checkers report that sky is blue and image of it being red has been digitally altered.'"
However the video does highlight the issue of deepfakes, which refers to the use of AI to create synthetic media — image, audio, video — in someone else’s likeness.
"While we have been hearing about deepfakes for a few years, we are now at an inflection point where the technology is at anyone’s disposal and it is only a matter of time before every one of us is a victim of deepfake technology," said Kartik Hosanagar, a professor of technology and digital business at the Wharton School of the University of Pennsylvania.
Last month, a fake video of Ukrainian President Volodymyr Zelenskyy appearing to tell his soldiers to lay down their arms and surrender the fight against Russia was circulated on social media.
Escalating Disinformation
A deepfake video of Russian President Vladimir Putin appeared online a short time later that depicted the former KGB-agent announcing Russia's surrender and telling soldiers to "go home while you're alive."
The actor Robert Pattinson was a recent target of a deepfake TikTok account. Pattison has said that he doesn’t have any social media accounts.
“Deepfakes can create illusions of an individual's presence and activities that did not occur in reality,” said Siwei Lyu, a professor of computer science and engineering and the director of the UB Media Forensic Lab at the University of Buffalo.
"As such, deepfakes can escalate the scale and danger of online disinformation. That is the reason why they can be very dangerous.”
Lyu cited a report last year that said the number of online deep fake videos doubles every six months.
"I think this is an underestimate and there will be many more in the coming years," he said.
Hosanagar said the technology can have many desirable uses, such as in films like "Forrest Gump," where Tom Hanks’ character meets JFK.
"But despite these positive use cases, the technology will more likely be in the news due to its ability to stoke civil and political unrest, savage reputations of individuals, and enable financial fraud," he added. "Every one of these potential scenarios has already played out and we are just at the start."
Technology Must Solve The Problem
As far as what to do about the threat of deepfakes, Hosanagar "technology has to solve the problem it has helped create."
"First, we need authentication systems that can verify the origin of images and videos, he said.
He said Microsoft has developed a prototype of a system named AMP: Authentication of Media via Provenance, which enables media content providers to create and assign a certificate of authenticity to their content.
"What this means for end users is that assurance regarding the authenticity of content they are viewing could be as simple as an icon (similar to the browser padlock icon) indicating that the media being consumed has not been tampered with," he said.
"While the creation of such authentication and provenance systems is a good first step, for these systems to be effective in practice they must be adopted broadly by content creators, which will take time."
In the meantime, Hosanagar said, the legal system could provide additional deterrents to stop the potential harm caused by deepfakes.
'All Hands on Deck!'
He said a number of laws are being proposed and adopted at the state and federal levels and he cited Virginia, which recently criminalized the non-consensual sharing of deepfake pornography by expanding a revenge porn law.
Another possibility is to ask online media platforms to do more in terms of authenticating content being shared on their platforms, he said.
"Overall, we need adoption of these kinds of laws, coupled with the adoption of authentication and provenance technologies amongst consumers and businesses," Hosanagar said.
Lyu agreed, saying that "fighting deepfakes is a community effort."
He said this includes research of more effective detection algorithms, adopting tighter screening and more effective tagging of falsified media by the social platforms.
The effort also requires legislation against the most offensive types of deepfakes, fast reaction and fact-checking by the public media, and due diligence online users.
"All hands on deck!" he said.
