AI Explored for Public Sector User Research
Emerging best practices for using AI in public sector user research were discussed in a recent online session. The tools are being piloted to automate tasks like survey analysis and sentiment extraction, accelerating the research process. However, experts caution that human judgment remains essential for qualitative depth and that algorithmic decisions in government services must be auditable and fair.
The European Union's AI Act, which came into force in 2024, establishes a comprehensive legal framework for artificial intelligence, classifying AI systems by risk to ensure they are safe and respect fundamental rights. This regulation has an extraterritorial reach, affecting any organization whose AI systems are placed on the EU market or whose outputs impact people within the EU. In response to such regulations, public sector bodies are increasingly adopting service design methodologies to ensure AI implementations are user-centered. This approach uses tools like journey mapping and co-creation with citizens to design services that are not only efficient but also genuinely meet user needs from the outset. The Scottish Government's "Scottish Approach to Service Design" is one example of embedding this philosophy at a national policy level. AI-powered tools are now capable of analyzing large volumes of qualitative data from citizen surveys, automatically categorizing topics, and detecting sentiment polarity (positive, negative, or neutral) using natural language processing. This allows government agencies to move from months-long manual analysis to identifying real-time issues, such as a broken appointment scheduler or a confusing permit workflow. Any AI-generated content or interface in the public sector must adhere to Web Content Accessibility Guidelines (WCAG). AI tools can introduce accessibility risks like poor color contrast or vague alternative text for images, meaning human oversight is essential to ensure that services are perceivable, operable, understandable, and robust for all citizens, including those using assistive technologies. To advance digital transformation, Portugal has focused on creating GovTech consortia and partnerships to accelerate the adoption of new technologies. This collaborative approach is one of several across Europe; Finland, for example, is developing "zero-touch" public services that are delivered automatically based on life events identified from trusted data sources, reducing the need for citizen interaction. Successfully integrating these tools requires building organizational design maturity, which involves moving design from a purely aesthetic function to a core part of strategic decision-making. A high level of design maturity fosters better cross-departmental collaboration and ensures that products, policies, and internal tools work together to create a consistent user experience. However, significant barriers to AI adoption in the public sector remain. The slow adaptation of legacy IT systems and rigid procurement processes often hinder the scaling of successful pilots. Successful AI integration ultimately depends on investing in the skills of civil servants and redesigning work processes to complement the new technology.
