Artificial intelligence (AI) has been a topic of interest for many countries as they explore ways to utilize it in government processes such as tax processing and welfare benefit determinations. While governments may see AI as a way to increase efficiency, a recent survey conducted in the UK revealed that citizens are not as enthusiastic about this technology as their officials. This lack of enthusiasm could pose serious risks for democracy, according to Alexander Wuttke, a researcher at the Ludwig Maximilian University of Munich in Germany.
Wuttke and his team surveyed approximately 1200 individuals in the UK, presenting them with scenarios where tasks were handled either by humans or AI. These scenarios included processing tax returns, assessing welfare applications, and determining bail eligibility for defendants. Participants were divided into groups, with some only being informed about the benefits of AI in government processes, while others were made aware of the potential risks associated with AI use. These risks included the opacity of AI decision-making processes, increased government reliance on AI that may become irreversible, and a lack of avenues for citizens to challenge AI decisions.
The survey results showed that when participants were informed about the risks of AI in government, there was a significant decline in trust in government and a feeling of loss of control. The percentage of individuals reporting a loss of democratic control in their government increased from 45% to over 81% in scenarios where AI was heavily relied upon for specific tasks. Additionally, the proportion of people advocating for less AI in government rose from under 20% in the baseline scenario to more than 65% in scenarios where both the benefits and risks of AI were presented.
Hannah Quay-de la Vallee from the Center for Democracy & Technology in Washington DC believes that democratic governments can utilize AI in responsible ways to maintain citizens’ trust. However, she notes that there have been few success stories of AI in government thus far, with several high-stakes failure cases already documented. For instance, automated processing of public benefits claims in certain US states resulted in thousands of individuals being wrongly charged for fraud, leading to severe financial consequences for some.
In conclusion, while AI has the potential to streamline government processes, it is crucial for officials to address the concerns and risks associated with its implementation. By ensuring transparency, accountability, and citizen participation in AI decision-making, governments can harness the benefits of AI technology without compromising democratic trust and legitimacy.
