AI deepfakes: tech could be used during election campaigns to incite hatred or spread misinformation on web

A study has raised concerns about the potential misuse of artificial intelligence-generated deepfakes in political contexts, particularly in the lead-up to the general election.

The research, conducted by The Alan Turing Institute’s Centre for Emerging Technology and Security (Cetas), highlights the risk of deepfakes being employed to fabricate false political endorsements or to deliberately confuse voters, undermining trust in the electoral process.

While there is currently limited evidence of direct impact on election outcomes, the researchers warned that there were early signs of damage to the broader democratic system.

This includes the potential for deepfakes to sow confusion among voters, and for AI to be exploited for the purposes of inciting hatred or spreading misinformation online.

Cetas has urged regulatory bodies such as Ofcom and the Electoral Commission to take proactive measures to address the growing threat of AI manipulation in public discourse.

It’s a big year for politics in the UK - stay informed with our Politics newsletter

The research team warned that currently, there was “no clear guidance” on preventing AI being used to create misleading content around elections.

Some social media platforms have already begun labelling AI-generated material in response to concerns about deepfakes and misinformation, and in the wake of a number of incidents of AI being used to create or alter images, audio or video of senior politicians.

In its study, Cetas said it had created a timeline of how AI could be used in the run-up to an election, suggesting it could be used to undermine the reputation of candidates, falsely claim that they have withdrawn or use disinformation to shape voter attitudes on a particular issue.

The study also said misinformation around how, when or where to vote could be used to undermine the electoral process.

Sam Stockwell, research associate at the Alan Turing Institute and the study’s lead author, said: “With a general election just weeks away, political parties are already in the midst of a busy campaigning period.

“Right now, there is no clear guidance or expectations for preventing AI being used to create false or misleading electoral information. That’s why it’s so important for regulators to act quickly before it’s too late.”

Dr Alexander Babuta, director of Cetas, said: “While we shouldn’t overplay the idea that our elections are no longer secure, particularly as worldwide evidence demonstrates no clear evidence of a result being changed by AI, we nevertheless must use this moment to act and make our elections resilient to the threats we face.

“Regulators can do more to help the public distinguish fact from fiction and ensure voters don’t lose faith in the democratic process.”