The Alan Turing Institute, has released a report indicating that AI-powered disinformation campaigns did not meaningfully affect the results of recent elections in the United Kingdom and Europe. The study, conducted by researchers from the Centre for Emerging Technology and Security (CETaS), identified only a handful of viral cases involving AI disinformation or deepfakes during these electoral events.
Sam Stockwell, lead author and Research Associate at The Alan Turing Institute, stated, "Echoing our previous report, there remains no evidence AI has impacted the result of an election, but we remain concerned about the persistent erosion of confidence in what is real and what is fake across our online spaces."
The report's findings reveal that only 16 confirmed viral instances of AI disinformation or deepfakes were identified during the UK general election, with a mere 11 cases found in the EU and French elections combined. Despite these reassuring numbers, the study highlighted new challenges posed by realistic parody or satirical deepfakes, which, while intended as humour, can sometimes include misleading election claims that voters might interpret as factual.
A concerning trend emerged in the targeting of politicians, particularly women, with deepfake pornography smears. This practice raises serious concerns about personal wellbeing and potential damage to professional reputations. Additionally, the researchers found evidence suggesting that some voters had difficulty distinguishing between legitimate political content and AI-generated material, potentially eroding public confidence in online information beyond the election context.
However, the report also noted positive applications of AI in the democratic process. These include climate campaigners using clearly labeled AI-generated parodies to amplify environmental issues, and fact-checkers employing AI to scrutinise political claims more efficiently.
The creation and circulation of deepfakes and AI-enabled disinformation were attributed to both domestic and state-sponsored groups. While most sharing of fake content was done by members of the public without explicit intention to undermine elections, researchers found some evidence of political candidates sharing deepfakes, particularly in the EU and French elections. The study also detected signs of interference from groups linked to Russia across all three elections, although this interference did not significantly impact the results.
Despite the reassuring findings regarding election outcomes, the report underscores the need for continued vigilance. Stockwell advised, "It's right that people are sceptical about the information they see, but by ensuring we use reliable sources, and cross-reference them with others, we can all be more confident about getting the information we need to inform our decisions at the ballot box."
As AI technology continues to evolve, the challenge for regulators and policymakers will be to strike a balance between countering disinformation and protecting free speech, including the role of satire in political discourse. The Alan Turing Institute's ongoing research aims to provide insights that can help strengthen future election resilience in the face of emerging AI-related threats. Their findings suggest that while AI disinformation had no significant impact on recent UK, French, or EU election results, the potential for voter confusion and targeted attacks on politicians remains a concern. The study underscores the importance of media literacy and critical thinking in navigating the complex landscape of online political information.