Making AI Safe for Democracy

Recent years have seen immense technological advances in the fields of Artificial Intelligence (AI) and machine learning. Crucially, AI and machine learning rely on data. Our contemporary societies are also hugely data-driven: citizenship and democracy itself are now heavily reliant on data. Without data neither the modern state nor a modern economy would be able to function. As a result, AI is not just the domain of technologists, but a powerful tool in the hands of policy-makers, industry, the media, and citizens themselves.

The event looks clearly at the impact that AI is having on contemporary democracy, and the way in which the social sciences can help ensure machine learning is used intelligently to support, rather than degrade, key values and practices of democracy.

The development of AI poses many questions for the practice and theory of democracy. To what extent, for instance, should polling agencies and political parties be allowed access to large data sets in order to model an individual's voting intention? Should, in trying to prevent political radicalisation, security agencies use sophisticated machine learning to analyse social media accounts? To what extent, if any, can the democratic state become 'digitised' - and would this make it more accountable? Who oversees the aggregation and utilisation of large data sets? The social and political sciences have an important role to play in answering these and related questions.

Organised by the Department of Politics at the University of Surrey, our inter-disciplinary event brings together researchers from across political science in order to address how AI can be made safe for democracy. Using many real-life examples, such as the 2016 EU referendum campaigns, the event showcases the importance of political science for our understanding of the challenges posed by AI. Talks by Surrey academics include the following topics: human rights and AI; political campaigning and AI; the use of virtual environments in democracies; the rule of law and AI; and the implications of AI for gender stereotypes and beyond.

The event is being held in the Lecture Room at 10-11 Carlton House Terrace, London. SW1Y 5AH.

Event timetable:

6.30pm-8.00pm : Panel discussion followed by Q&A

8.00pm-9.00pm: Reception