Paper Number
ICIS2026-2158
Paper Type
Completed
Abstract
Information Systems (IS) research has predominantly treated regulation as stable and top-down: rules are set, and actors comply. This framing is inadequate for governing AI. Rapid technological change outpaces regulatory updates, rendering frameworks obsolete. Adaptive regulation addresses this by designing frameworks that learn from the actors they govern. The central mechanism is regulatory learning: generating and integrating evidence from practice to trigger updates. Drawing on the AI Act, where adaptive ambitions are explicitly codified in law, we show that the Act's adaptive infrastructure remains largely incomplete. The metrics, knowledge infrastructures, and coordination required to make regulatory learning operational do not yet exist. This gap opens three paths for IS research: (1) designing knowledge infrastructures that support regulatory learning; (2) empirically studying regulatory learning under ambiguity and resource constraints; and (3) engaging in IS policy research.
Recommended Citation
Attanayake, Charuni; Lanamäki, Arto; and Väyrynen, Karin, "Adaptive AI Regulation Beyond Aspiration: Regulatory Learning and the AI Act" (2026). ICIS 2026 Proceedings. 3.
https://aisel.aisnet.org/icis2026/policy/policy/3
Adaptive AI Regulation Beyond Aspiration: Regulatory Learning and the AI Act
Information Systems (IS) research has predominantly treated regulation as stable and top-down: rules are set, and actors comply. This framing is inadequate for governing AI. Rapid technological change outpaces regulatory updates, rendering frameworks obsolete. Adaptive regulation addresses this by designing frameworks that learn from the actors they govern. The central mechanism is regulatory learning: generating and integrating evidence from practice to trigger updates. Drawing on the AI Act, where adaptive ambitions are explicitly codified in law, we show that the Act's adaptive infrastructure remains largely incomplete. The metrics, knowledge infrastructures, and coordination required to make regulatory learning operational do not yet exist. This gap opens three paths for IS research: (1) designing knowledge infrastructures that support regulatory learning; (2) empirically studying regulatory learning under ambiguity and resource constraints; and (3) engaging in IS policy research.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
21_Policy_and_regulations