AIF Insights No. 12 (2026) | Navigating Regulatory Impact Assessment in the Age of AI

This article examines the evolution of Regulatory Impact Assessment (RIA) from a manual, economic-centric “box-ticking” exercise into a multi-dimensional, AI-driven model of “evidential reasoning”. In the current “Digital Age,” governments are transitioning from being mere regulators of AI to active users of the technology to enhance rule-making through “Agile Regulatory Governance” and “Rules as Code” (RaC). By leveraging AI for text mining, sentiment analysis in public consultations, and automated identification of administrative burdens, policymakers can overcome information overload and create more inclusive, data-driven interventions. However, to mitigate risks such as automation bias and data discrimination, the author argues that a strategic framework must be adopted that prioritizes transparency, multidisciplinary skills, and a strict “human-in-the-loop” approach to ensure technology supports rather than replaces accountable human judgment