This whitepaper was produced by Professor Josiah Ober and Professor John Tasioulas.
Executive Summary
Many today take the view that the AI technological revolution is creating a radically new reality, one that demands a corresponding upheaval in our ethical thinking. This can generate a sense of helplessness in the face of rapid technological advances. But the idea that we start from an ethical blank slate in addressing the challenges and opportunities of this transformative technology is a fallacy.
The key contention of this white paper is that the basic approach to ethics developed by the 4th Century BC Greek philosopher, Aristotle, offers the most compelling framework for addressing the challenges and opportunities of Artificial Intelligence today.
Among the key elements of an Aristotelian approach to AI ethics explored in the white paper are the following:
A truly ‘human centred’ approach to the ethics of AI, one that conceives both human flourishing and human morality as rooted in our nature as human beings whose fulfilment depends on the exercise of capacities for rationality, social engagement, and communication. The Aristotelian approach constantly foregrounds the question of the good for which AI systems are developed and deployed and does not conceive of ethics as in competition with technological advance.
A richer conception of ethics than the dominant ethical theories in the discourse of AI: on the one hand, approaches grounded in the fulfilment of preferences or the maximisation of wealth; on the other hand, approaches based on human rights law. The former are focussed on considerations that are not ultimate values; the latter are incomplete, failing to recognise that considerations besides rights, such as virtues and the common good, are essential.
An emphasis on the powerful nexus between ethics and politics, since human beings can only flourish in communities, and the importance of democracy (Aristotle’s notion of citizens as ‘ruling and being ruled in turn’) and liberalism (given the importance of free choice in the cultivation and exercise of the Aristotelian virtues) as political values.
The idea of AI systems as ‘intelligent tools’ to be deployed in order to advance the flourishing of individuals and communities, a focus very different from the dominant objective of the tech industry, which is to create Artificial General Intelligence that replicates human intellectual capacities across their entire spectrum.
An elucidation of how Aristotle’s own doomed attempt to justify the existence of a class of ‘natural slaves’ serves as a cautionary illustration of the perils of seeking to instrumentalise beings with intellectual capabilities comparable to humans.
The development of the idea of AI systems as ‘intelligent tools’ in two domains – that of work, where their fundamental role is to augment the exercise of valuable human capacities, rather than to replace systematically human endeavour, and that of democracy, where AI tools can play an important role in enabling a genuinely deliberative and participatory democracy to operate at the scale of the modern state.
The guideposts that the Aristotelian framework offers in regulating AI, overcoming unduly restrictive regulatory rubrics like ‘safety’ and ‘existential risk’, and providing a more comprehensive regulatory framework than the frameworks deployed by the world’s three digital empires: state-based (China), market-based (US), and rights-based (EU).
A template for addressing the challenge of how international co-operation might be achieved in regulating AI globally without international law and institutions intruding unduly into the proper sphere of decision that belongs to states.
A case for a novel human right for the age of AI: a qualified right to a human decision that has the effect of prohibiting certain decisions from being made by AI systems, or else allowing opt-outs or appeals from such decisions where they are permissible.
The authors gratefully acknowledge support in developing this paper from the Patrick J McGovern Foundation, the World Human Forum, the Cosmos Institute, Demokritos, Oxford University Institute for Ethics in AI, and Stanford University Human-Centred Artificial Intelligence.
Download the full whitepaper below.