The Source of Unexplored Opportunities or an Unpredictable Risk Factor?

Abstract

The Collingridge dilemma or ‘dilemma of control’ presents a problem at the intersection of law, society and technology. New technologies can still be influenced, whether by regulation or policy, in their early stage of development, but their impact on society remains unpredictable. In contrast, once new technologies have become embedded in society, their implications and consequences are clear, but their development can no longer be affected. Resulting in the great challenge of the pacing problem – how technological development increasingly outpaces the creation of appropriate laws and regulations. My paper examines the problematic entanglement and relationship of Artificial Intelligence (AI) and a key aspect of the rule of law, legal certainty. AI is our modern age’s fastest developing and most important technological advancement, a key driver for global socio-economic development, encompassing a broad spectrum of technologies between simple automation and autonomous decision-making. It has the potential to improve healthcare, transportation, communication and to contribute to climate change mitigation. However, its development carries an equal amount of risk, including opaque decision-making, gender-based or other kinds of discrimination, intrusion into private lives and misuse for criminal purposes. The transformative nature of AI technology impacts and challenges law and policymaking. The paper considers the impact of AI through legal certainty on the rule of law, how it may undermine its various elements, among others foreseeability, comprehensibility and clarity of norms. It does so by elaborating on AI’s potential threat brought on by its opacity (‘black box effect’), complexity, unpredictability and partially autonomous behaviour, which all can impede the effective verification of compliance with and the enforcement of new as well as already existing legal rules in international, European and national systems. My paper offers insight into a human-centric and risk-based approach towards AI, based on consideration of legal and ethical questions surrounding the topic, to help ensure transparency and legal certainty in regulatory interventions for the benefit of optimising efficiency of new technologies as well as protecting the existing safeguards of legal certainty.