The Alignment Problem: Machine Learning and Human Values

As artificial intelligence (AI) systems are becoming increasingly autonomous and will soon be able to make decisions on their own about what to do, AI researchers have started to talk about the need to align AI with human values. The AI ‘value alignment problem’ faces two kinds of challenges—a technical and a normative one—which are interrelated. The technical challenge deals with the question of how to encode human values in artificial intelligence. The normative challenge is associated with two questions: “Which values or whose values should artificial intelligence align with?” My concern is that AI developers underestimate the difficulty of answering the normative question. They hope that we can easily identify the purposes we really desire and that they can focus on the design of those objectives. But how are we to decide which objectives or values to induce in AI, given that there is a plurality of values and moral principles and that our everyday life is full of moral disagreements? In my paper I will show that although it is not realistic to reach an agreement on what we, humans, really want as people value different things and seek different ends, it may be possible to agree on what we do not want to happen, considering the possibility that intelligence, equal to our own, or even exceeding it, can be created. I will argue for pluralism (and not for relativism!) which is compatible with objectivism. In spite of the fact that there is no uniquely best solution to every moral problem, it is still possible to identify which answers are wrong. And this is where we should begin the value alignment of AI.

Download Full-text

AI Safety and Reproducibility: Establishing Robust Foundations for the Neuropsychology of Human Values

10.31219/osf.io/7z63x ◽

2017 ◽

Cited By ~ 1

Author(s):

Gopal P. Sarma ◽

Nick J. Hay ◽

Adam Safron

Keyword(s):

Artificial Intelligence ◽

Human Values ◽

Systematic Effort ◽

Alignment Problem ◽

The Creation ◽

Ai Safety

We propose the creation of a systematic effort to identify and replicate key findings in neuroscience and allied fields related to understanding human values. Our aim is to ensure that research underpinning the value alignment problem of artificial intelligence has been sufficiently validated to play a role in the design of AI systems.

Download Full-text

AI Ethics and Value Alignment for Nonhuman Animals

Philosophies ◽

10.3390/philosophies6020031 ◽

2021 ◽

Vol 6 (2) ◽

pp. 31

Author(s):

Soenke Ziesche

Keyword(s):

Human Values ◽

Nonhuman Animals ◽

Alignment Problem ◽

Ai Ethics ◽

Do So ◽

Ai Safety

This article is about a specific, but so far neglected peril of AI, which is that AI systems may become existential as well as causing suffering risks for nonhuman animals. The AI value alignment problem has now been acknowledged as critical for AI safety as well as very hard. However, currently it has only been attempted to align the values of AI systems with human values. It is argued here that this ought to be extended to the values of nonhuman animals since it would be speciesism not to do so. The article focuses on the two subproblems—value extraction and value aggregation—discusses challenges for the integration of values of nonhuman animals and explores approaches to how AI systems could address them.

Download Full-text

IR 4.0. IN PARLIAMENT: CONCEPTUALISING THE APPLICATION OF ARTIFICIAL INTELLIGENCE AND MACHINE LEARNING IN THE PARLIAMENT OF MALAYSIA’S PARLIAMENTARY QUESTIONS

International Journal of Law, Government and Communication ◽

10.35631/ijlgc.5200010 ◽

2020 ◽

Vol 5 (20) ◽

pp. 124-137

Author(s):

Muthanna Saari

Keyword(s):

Artificial Intelligence ◽

Machine Learning ◽

Industrial Revolution ◽

Emerging Technologies ◽

Participant Observation ◽

Research Data ◽

Human Values ◽

Structured Interviews ◽

Fourth Industrial Revolution ◽

The Government

The Fourth Industrial Revolution (IR 4.0.) offers significant opportunities to humankind in revitalising human values through which the emerging technologies inevitably seam into daily societal life. Legislatures face ever-increasing challenges in fulfilling their duties in such a complicated society which subsequently entails complex legislations. Parliamentary questions (PQs) as one of the traditional tools utilised by parliamentarians provide a quintessential mechanism to achieve the oversight functions of parliament. However, there are still immense undiscovered potentials of PQs, yet many previous studies have not looked into the content of the questions and the consequences of the response to the conduct of governments. This paper set out to examine the usefulness of IR 4.0. namely, artificial intelligence (AI) and machine learning towards improving the efficiency, transparency, and accountability of parliament and the government. The research data of this exploratory and interpretative study is drawn from three main sources: literature studies, semi-structured interviews, and participant observation of the existing PQs processing in the Dewan Rakyat, Parliament of Malaysia. This study has found that generally, the approval of such technologies introduction to the parliamentary businesses is contingent upon its ability to capture complex considerations in the existing environment.

Download Full-text

Mind wandering as data augmentation: How mental travel supports abstraction

Behavioral and Brain Sciences ◽

10.1017/s0140525x1900311x ◽

2020 ◽

Vol 43 ◽

Author(s):

Myrthe Faber

Keyword(s):

Machine Learning ◽

Data Augmentation ◽

Mental Content ◽

Mind Wandering ◽

Theoretical Framework ◽

Important Addition

Abstract Gilead et al. state that abstraction supports mental travel, and that mental travel critically relies on abstraction. I propose an important addition to this theoretical framework, namely that mental travel might also support abstraction. Specifically, I argue that spontaneous mental travel (mind wandering), much like data augmentation in machine learning, provides variability in mental content and context necessary for abstraction.

Download Full-text