01. Abstract | Being ethically intelligent about AI

How might we put ethics at the core of the AI design process and give teams the tools they need to safely consider and align on long-term ethical values and design principles?

Abstract

The rise of AI is inevitable. The development of artificial intelligence has recently seen a sharp increase and has prompted words of caution from prominent figures in the field. While the benefits of AI systems and products are enormous, we are currently seeing the release of more innovations before we have fully explored the impact it has on our lives.

While we might be some way off developing human-destroying robots, the last eighteen months have seen a need to discuss the ethical impact of AI in the future and has sparked a surge for the need of ethically-minded conversations within design teams.

If teams that use design thinking within their AI projects can have ethics at the core of their design process, research suggests that talking about ethics throughout the entire project can help to bring ethical considerations to the forefront of their thinking. By not actively designing with ethics on the table, we run the risk of repeating the times in history where we did not consider ethics, either through further acts of war, the reinstatement of slave industries or bringing about health crises.

The purpose of this paper is to explore how AI design teams currently consider ethics when designing AI systems. Throughout this project, experience design practices are used to research and then develop an ethical toolkit for teams designing for AI that will allow them to explore and align on their ethical values. A prototype will be created from insights discovered through secondary research and interviews with experts and users with a focus on AI and ethics. Iterations will be made to the prototype based on feedback from users and findings will be concluded before future steps, and personal reflections on the project are outlined.

Terminology

Emerging Technology

A “relatively fast growing and radically novel technology characterised by a certain degree of coherence persisting over time and with the potential to exert a considerable impact on the socio-economic domain(s) … Its most prominent impact, however, lies in the future and so in the emergence phase is still somewhat uncertain and ambiguous”. (Rotolo, Hicks and Martin, p.1840)

Artificial Intelligence (AI)

In the context of this research paper, artificial intelligence should be read broadly as a catch-all term “for a large number of sub(fields)” (European Economic and Social Committee, 2017, p.4), including the likes of cognitive computing and algorithms, machine learning, augmented intelligence and AI robotics.

Ethics

Ethics are the focal point of this project and should be taken to mean “moral principles that govern a person’s behaviour or the conducting of an activity” (Oxford Dictionaries, n.d.) and to “analyse complex problems and make decisions” (Resnik, 2015) on “the grounds on which moral statements are made” (Stahl, 2012).

Teams

The teams referred to throughout this paper should be taken to mean multidisciplinary teams, which may consist of, but is not limited to, a number of designers, researchers, data scientists, engineers and anybody else directly involved in the design and development process for AI systems, products or services.

Designer

In this paper, the term designer is used broadly to refer to any one individual team member, whose role in the team is aiding in the design of AI products and systems.