Leveraging AI to enhance public sector audits: early detection of risks

A new report, 'Early Warning Signs in Public Sector Bodies,' published by the Committee on Standards in Public Life, highlights how AI and technology are transforming the effectiveness of internal audits in the public sector.

The Government Internal Audit Agency (GIAA), prominently featured in the report, demonstrates an innovative approach by utilising AI to conduct real-time data analysis against risk criteria. This technology-driven method significantly enhances the active identification of emerging issues, allowing for quicker resolution and more efficient mitigation of risks.

Harriet Aldridge, Chief Executive of GIAA, emphasised the critical role AI plays in modern audit practices. “A robust internal audit approach, supported by AI, should help identify issues upstream,” Aldridge explained. “Earlier resolution ensures better outcomes for the public, saving taxpayers’ money and leading to the more effective delivery of public services.”

The report underscores the growing importance of integrating advanced technologies into auditing processes to overcome traditional limitations such as delayed detection of issues and inefficient investigations. It also commends GIAA for its leadership in leveraging these technological tools to share insights and best practices across government entities through its cross-government Insight Programme.

Doug Chalmers, Chair of the Committee on Standards in Public Life, noted that the success of technological initiatives hinges on a culture of transparency and openness. "Organisations with strong leadership and open cultures are better equipped to leverage technology like AI to detect and address risks early," said Chalmers.

The report calls on public sector leaders to adopt and encourage the use of AI and other technologies, alongside fostering a supportive environment where employees are motivated to proactively identify and communicate potential concerns, driving improved public outcomes and efficiency.