An alarming surge of over 100 deep fake video advertisements impersonating UK Prime Minister Rishi Sunak on Facebook has raised serious concerns about the vulnerability of social media platforms to AI manipulation. Here’s the full story.
AI Threat to Democracy
AI has long been proclaimed to be the next generation of technology, a way of integrating computer and human intelligence to make all our lives easier. However, recent research has discovered a shocking and unsettling other use for the technology. There has been a flood of deep fake videos released online which appear to contain uncannily accurate versions of Prime Minister Rishi Sunak on Facebook.
These deep fake videos have been promoted as paid adverts on Facebook, which has deeply unsettling possible consequences of AI manipulation in the upcoming UK election.
The ease of creating and distributing these deep fakes reveals a significant escalation in the use of fake or manipulated content, with unsettling consequences for the democratic process, not just in the UK but in every democracy around the world.
The Spread of a Digital Virus
Marcus Beard, a former official in the UK government tasked with combating conspiracy theories spreading during the Covid-19 pandemic and founder of Fenimore Harper, a communications company, sheds light on the alarming surge in deep fake video ads.
“With the advent of cheap, easy-to-use voice and face cloning, it takes very little knowledge and expertise to use a person’s likeness for malicious purposes,” stated Beard. According to Beard, the ads in question, which violate several of Facebook’s policies, had a staggering potential reach online of up to 400,000 unwitting people.
The unprecedented scale and ease with which potential bad actors could utilize this technology has raised a number of serious questions about the vulnerability of social media platforms and their ability to quickly spread AI-generated falsehoods.
Global Reach
The research by Beard and his team has shown that these deep fake adverts originated from over 23 countries, as varied and widespread as the US, Turkey, Malaysia, and the Philippines. The amount spent to gain such extensive reach with the AI-generated videos was shockingly low, with £12,929 spent on 143 adverts.
This amount, though small in terms of genuine political advertising, had an international dimension which is worrying many who work in governments and the regulators they employ.
The research shows that these deepfake manipulation campaigns were not merely about impersonating political figures but actually conjured up elaborate alternative scenarios to deceive anyone who might see them. One advert featured faked footage of a BBC newsreader who falsely claimed that Sunak had earned substantial earnings from a project that was originally intended to benefit UK citizens.
However, the video did not stop there, continuing to a seemingly plausible news story, also faked, which suggested that Elon Musk had launched an app for collecting transactions generated from stock market trades.
The video then finished with the Sunak AI deep fake apparently promoting the app, endorsing the government’s decision to test the application, which was actually a scam investment scheme set up to defraud users.
Social Media Contagion
Marcus Beard, the mind behind Fenimore Harper, stressed the problem with such scam campaign videos being so easily disseminated by social media sites, such as Facebook, when he stated, “Unfortunately, this problem is exacerbated by lax moderation policies on paid advertising.
These adverts are against several of Facebook’s advertising policies. However, very few of the ads we encountered appear to have been removed.”
Somewhat unsurprisingly, considering their failure to remove the fake videos, Meta, the parent company of Facebook, has yet to respond to the findings.
However, a UK government spokesperson stated, “We are working extensively across government to ensure we are ready to rapidly respond to any threats to our democratic processes through our defending democracy taskforce and dedicated government teams.
Our Online Safety Act goes further by putting new requirements on social platforms to swiftly remove illegal misinformation and disinformation – including where it is AI-generated – as soon as they become aware of it.”
The BBC’s Counter Attack
The BBC, one of the main targets in the series of deep fake ads analyzed by the researchers, emphasized its commitment to countering the spread of misinformation.
A spokesperson noted, “In a world of increasing disinformation, we urge everyone to ensure they are getting their news from a trusted source. We launched BBC Verify in 2023 to address the growing threat of disinformation – investing in a highly specialized team with a range of forensic and open source intelligence (OSINT) to investigate, factcheck, verify video, counter disinformation, analyze data and explain complex stories.
We build trust with audiences by showing how BBC journalists know the information they are reporting, and offer explainers on how to spot fake and deepfake content. When we become aware of fake BBC content we take swift action.”
Regulatory Efforts
Regulators, including the Electoral Commission in the UK, have expressed deep concerns that time may be running out to implement the comprehensive changes necessary to protect and safeguard the electoral system against the ever-advancing march of AI.
Recent legislation from 2022 has required an “imprint” on digital campaign material, which, it is hoped, will be a useful step in ensuring transparency and validity of online campaigns.
However, the ease with which these most recent videos were disseminated and spread leaves a worrying precedent and highlights the need for government and regulators to work together to ensure the integrity of the democratic process.
The delicate balance between freedom of expression and preventing the abuse of advanced technologies remains a central concern as the world grapples with misinformation and digital deception in the age of AI.
The post Over 100 Deepfake Ads Misuse UK Prime Minister’s Image on Facebook first appeared on Pulse365 Limited.
Featured Image Credit: Shutterstock / Alessia Pierdomenico.