Balancing AI Speed and Control, Patrick Zalda

Patrick Zalda
Research Student, DBA Program
Heriot-Watt University
UK

Purpose: There are many reasons for the existence and further development of AI. However, with the rapid technical development of AI, ‘guard rails’ are needed to ensure the safe and human-centred use of AI.
Design/methodology/approach: The research conducts a Delphi study in which experts from professional, academic and political environments are interviewed in iterative surveys. The aim of the study is to determine to what extent ethical aspects can be integrated into the implementation process of AI from the perspective of experts in AI.
Findings: The initial suggestions from the research are that organisations are not well aware of the ethical issues themselves and even where some do have a strong awareness of such issues they have not really considered how to take ethical aspects of AI into account.
Added value / practical implications: Guidelines for the safe and human-centred use of AI are new. Companies that use or produce AI will be able to benefit from the full research results, especially in the EU where the AI Act is already in force – compliance with this Act will be helped by several strands of this ongoing research.
Keywords: Artificial Intelligence, Ethics, Delphi Study, Risk Management, Governance, EU AI Act.