Artificial intelligence and universal values / by Jay Friedenberg.

Format:

Book

Author/Creator:

Friedenberg, Jay, author.

Language:

English

Subjects (All):

Artificial intelligence.

Ethics.

Physical Description:

xiii, 153 pages ; 22 cm

Edition:

First edition.

Place of Publication:

United Kingdom : Ethics International Press Ltd, 2024.

Language Note:

In English

Summary:

The field of value alignment, or more broadly machine ethics, is becoming increasingly important as artificial intelligence developments accelerate. By 'alignment' we mean giving a generally intelligent software system the capability to act in ways that are beneficial, or at least minimally harmful, to humans. There are a large number of techniques that are being experimented with, but this work often fails to specify what values exactly we should be aligning. When making a decision, an agent is supposed to maximize the expected utility of its value function. Classically, this has been referred to as happiness, but this is just one of many things that people value. In order to resolve this issue, we need to determine a set of human values that represent humanity's interests. Although this problem might seem intractable, research shows that people of various cultures and religions actually share more in common than they realize. In this book we review world religions, moral philosophy and evolutionary psychology to elucidate a common set of shared values. We then show how these values can be used to address the alignment problem and conclude with problems and goals for future research. The key audience for this book will be researchers in the field of ethics and artificial intelligence who are interested in, or working on this problem. These people will come from various professions and include philosophers, computer programmers and psychologists, as the problem itself is multi-disciplinary.

Contents:

Intro

Introduction

The Intelligence Explosion

Values and Ethics

Chapter Previews

Chapter 1 The Benefits and Risks of Intelligence

Intelligence

Artificial Intelligence

Large Language Models

Limitations of Current AI Systems

Artificial General Intelligence

Artificial Superintelligence

AI Risks

Examples of AI Risk

AI Safety

Conclusion

Chapter 2 Artificial Intelligence and The Role of Values

Types of Value

Arguments and Counterarguments on the Concept of Value

Value Convergence

The Value Alignment Problem

Value Alignment Methods

Goals of Value Alignment

Principles for Value Alignment

Chapter 3 Ethics and Artificial Intelligence

Ethics and Morality

Meta-ethics and AI

Normative Ethics and AI

Virtue Ethics and AI

Applied Ethics

Philosophy and Machine Ethics

Taxonomies of Moral Agents

Autonomous Moral Agents

Real World Deployment of Autonomous Moral Agents

Chapter 4 Psychological Approaches to AI and Values

Evolutionary Theories of Value

Psychological Theories of Value

Value Theories

Literature Studies

Survey Studies

Chapter 5 The Ecological Approach and AI

Ecological Value and AI

AI Agent Systems

An Ecological Theory of AI Safety

Networks

Foodwebs

Environments

Evolution

Populations

Species

Communities

Species Interactions

Species Invasion and Defense

Species Competition

Species Deletion and Coherence

Human Societal Applications

Chapter 6 Ethical Problems in AI and Human Values

Automation and Human Employment

Military AI

Autonomous Vehicles

Smart Cities and AI

Surveillance and AI

AI and Legal Systems

Healthcare and AI

Teaching and AI

AI, Creativity and Art

AI and Social Relationships.

Human Values and Ethical Problems

Codes and Standards

The ART of AI

Chapter 7 Universal Values and AI

The Case Against Universal Values

The Case for Universal Values

A Metatheory of Universal Value

The Five Levels

Problems with the Universal Values Approach

How Advanced AI Might Create Better Ethics

Supra-human Values

The Future of Value

General Conclusion

References.

Notes:

Includes biblographical references (pages 128-153).

ISBN:

9781804416075