Jacinda Ardern’s deepfake clip was shared widely. She was the latest target of a disturbingly advanced type of technology.
When a video showing New Zealand Prime Minister Jacinda Ardern abusing drugs surfaced on social media in recent months, experts quickly dismissed it as fake.
The video, which was viewed and shared thousands of times, showed a woman smoking as a Crack Pipe.
The PM’s face was superimposed using artificial intelligence. But this video made for YouTube was quite convincing to many who shared it.
It was the latest example of how disturbingly authentic-looking videos can blur the lines between reality and fiction.
Experts say such videos created using deep-fake technology are becoming so sophisticated that it will soon be “nearly impossible to detect a fake photo or video with the naked eye”.
They say its use would extend to identity theft, sexual abuse, damage to reputation and harassment, “military deceit” and “erosion of trust in institutions and fair election processes”.
Loss of trust in democratic institutions
Australian Strategic Policy Institute Latest Reports on Technology, Title armed deep fake, takes a deeper look at where the problem is headed.
“Deep fakes will pose the most risks when combined with other technologies and societal trends: they will increase cyberattacks, accelerate the spread of online propaganda and disinformation, and undermine trust in democratic institutions,” the report said.
The authors say the “Russian model” of disinformation—sharing large amounts of propaganda—will benefit the most in years to come.
“Online propaganda is already a significant problem, especially for democracies, but deep counterfeiting will reduce the cost of engaging in large-scale information warfare and broaden the range of actors able to engage,” the report said. written.
“Today, propaganda is largely generated by humans, such as China’s ’50-centre’ and Russian ‘troll farm’ operators. However, improvements in deep counterfeiting technology, particularly text-generation tools, allow humans to ‘ out of the loop can help.
“The main reason for this is not because deep counterfeits are more authentic than human-generated content, but that they can produce ‘good enough’ content faster and more economically than current models for information warfare. can.
“Deep counterfeiting technology will be a particular value-add to the so-called Russian model of propaganda, which emphasizes the amount and intensity of propaganda over plausibility and consistency to overwhelm, disorient and divide a target.”
‘Nearly impossible to recognize with the naked eye’
Australia’s eSafety Commissioner, dedicated to keeping Australians safe online, notes in her position statement on deep counterfeiting that “advancements in artificial intelligence and machine learning have taken the technology even further, making it increasingly vulnerable to such threats.” allows to generate content that is extremely realistic, nearly impossible to detect with the naked eye and difficult to debunk”.
“Indeed, the field is developing so rapidly that deepfake material can be generated without the need for any human supervision, called a recycled generative adversarial network,” the statement said.
“Deep fakes have the potential to cause significant harm. To date, they have been used to create fake news, false porn videos, and malicious hoaxes, usually targeting famous people such as politicians and celebrities. Potential More specifically, deepfakes can be used as a tool for identity theft, extortion, sexual abuse, reputation damage, ridicule, intimidation and harassment.
The position statement said the deep fakes included “increased potential for fraud, propaganda and propaganda, military deception and erosion of trust in institutions and fair election processes”.
real-world use that made an impact
Perhaps the most widely shared deeply fake video comes from the United States, where House Speaker Nancy Pelosi is the victim of a horrific attempt to undermine her credibility,
The video showed Ms Pelosi speaking on camera, but the video and audio were rigged to show her intoxicated.
The video has been shared thousands of times and has been viewed millions of times.
Donald Trump’s personal lawyer at the time, Rudy Giuliani, tweeted a link to the video, giving it more credibility.
“What’s Wrong With Nancy Pelosi?” She wrote. “His speech pattern is bizarre.”
Facebook founder Mark Zuckerberg was forced to take action after a deeply fake video showed him mocking Internet users.
Video posted on Instagram (who owns Facebook), falsely portrayed the billionaire as saying: “Imagine this for a second: one man, with the stolen data of billions of people, with total control of all their secrets, their lives, their future. With.”
The video is still online today.
Experts say that as deeply faked videos become more sophisticated, many leave behind telltale signs of doctoring.
These signs include blurring or pixelation, especially around the mouth and eyes, badly synced sounds, slurs, changes in lighting, story gaps and irregular blinking.