Democracy Insight has today launched our new AI Transparency campaign alongside the publication of our Artificial Intelligence (AI) Policy.

The campaign has been created in response to the rapid growth of artificial intelligence within politics, campaigning and public life. While AI has the potential to improve communication, increase accessibility and help people engage with democratic processes, it also raises important questions about transparency, accountability and trust.

Recent advances in AI technology mean it is now possible to create highly realistic images, videos and audio recordings that can be difficult, although thankfully not impossible, to distinguish from genuine content. These developments offer exciting opportunities, but they also create risks, particularly when used within political campaigns and democratic debate.

Voters should know when AI is used

Our new AI Transparency campaign argues that voters have a right to know when political content has been generated, significantly modified or influenced by artificial intelligence. They also need to understand how political campaigners, parties and candidates are using AI to support their content creation and comments.

We are calling for clear disclosure standards, stronger protections against deceptive deepfakes and the adoption of AI policies and codes of best practice by political parties and campaign organisations.

The campaign is not anti-technology. Artificial intelligence is here to stay and has the potential to bring significant benefits. However, we believe innovation must be accompanied by transparency if public trust is to be maintained.

Democracy Insight's AI policy

Alongside the campaign, Democracy Insight has also published our own AI Policy. The policy sets out how artificial intelligence is used within the organisation and the safeguards that are in place.

At Democracy Insight, we use AI to support research, idea generation, drafting and editing. However, all content published on the website is subject to human review, editorial judgement and final approval before publication. AI assists our work, but it does not replace accountability.

If an article, campaign page or piece of analysis appears on Democracy Insight, you can be reassured it has been carefully read, reviewed and signed off before publication.

We believe organisations should be open about how they use AI, particularly when operating in areas connected to politics, public affairs and democratic engagement. By publishing our own policy, we hope to lead by example and contribute positively to an important conversation about the future relationship between technology and democracy.