AI-generated ‘deepfake’ audio clips of both London mayor Sadiq Khan and leader of the opposition Sir Keir Starmer have circulated on social media. A faked robocall impersonating Joe Biden was sent to voters before a primary election. The number of AI-enhanced images of politicians is increasing.

In a year when over two billion people in more than 50 countries will vote in elections, the use of AI technology – more widely accessible than ever – and disinformation could have a major impact on trust and integrity.

So what can be done to tackle this growing problem? What can political parties, government, media companies and civil society do to mitigate the risks of AI and ensure electoral integrity? And what steps can be taken ahead the elections of 2024 and then in the longer term.

We were joined by an expert panel, including:

Professor Joe Burton, Professor of International Security in the Department of Politics, Philosophy and Religion at Lancaster UniversityLouise Edwards, Director of Regulation and Digital Transformation at the Electoral CommissionChris Morris, Chief Executive of Full FactSimon Staffell, Director of Government Affairs at Microsoft

The event was chaired by Jill Rutter, Senior Fellow at the Institute for Government.

We would like to thank Lancaster University for kindly supporting this event.