Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

The Artificial Moral Advisor. The "Ideal Observer" Meets Artificial Intelligence.

Alberto Giubilini1, Julian Savulescu2

  • 11Oxford Martin School, University of Oxford, Oxford, UK.

Philosophy & Technology
|July 6, 2018
PubMed
Summary

This article introduces the Artificial Moral Advisor (AMA), a computational system designed to assist people in making better moral choices. By combining classic philosophical concepts with modern technology, the AMA offers impartial guidance that respects individual values while helping users overcome common cognitive biases.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Best Interest, Basic Interest, and Parental Autonomy in Child Vaccination Choices. On Roland Pierik and Marcel Verweij's <i>Inducing Immunity?</i>

Public health ethics·2026
Same author

Human Identity and Biotechnologies.

Bioethics·2026
Same authorSame journal

Privacy and Human-AI Relationships.

Philosophy & technology·2026
Same author

An Ethical Analysis of Public Attitudes towards Controlled Human Infection Studies in Singapore: Acceptability and Payment.

Asian bioethics review·2026
Same author

Additive Manufacturing of a PA11 Prototype Fabricated via Selective Laser Sintering for Advanced Industrial Applications.

Polymers·2025
Same author

Medical humanities, bioethics, and the (im)possibility of interdisciplinarity. Introduction to the special issue on the medical humanities in the 21<sup>st</sup> century.

Monash bioethics review·2025

Area of Science:

  • Applied ethics and artificial moral advisor research within cognitive science
  • Human-computer interaction studies in moral philosophy

Background:

Current ethical frameworks struggle to address the inherent limitations of human decision-making processes during complex moral dilemmas. No prior work had resolved how computational systems might bridge the gap between abstract philosophical ideals and practical human application. Scholars have long debated the feasibility of creating objective, impartial agents capable of providing moral guidance. That uncertainty drove the exploration of integrating classical ethical theories into modern digital architectures. Traditional models often failed to account for individual autonomy while attempting to maintain consistency in judgment. This gap motivated the development of systems that balance universal principles with personal values. Researchers have sought ways to mitigate human cognitive biases without suppressing the beneficial aspects of emotional intuition. The proposed framework aims to reconcile these competing demands through a novel technological approach.

Purpose Of The Study:

Keywords:
Artificial moral advisorIdeal observer theoryMoral artificial intelligenceMoral enhancementMoral psychologymoral philosophycomputational ethicsreflective equilibriumcognitive bias mitigation

Frequently Asked Questions

The Artificial Moral Advisor functions by applying a quasi-relativistic version of the ideal observer theory. It provides consistent, disinterested guidance while incorporating the user's personal principles, unlike the strictly absolutist approach found in traditional philosophical models.

The system utilizes the ideal observer concept, originally formulated by Roderick Firth. This framework allows the advisor to remain dispassionate and impartial, serving as a stable reference point for users navigating difficult ethical choices.

A non-absolutist design is necessary because it allows the system to respect the unique values and principles of the human agent. This ensures that the advice provided remains relevant and acceptable to the individual user.

The advisor acts as a cognitive tool that processes information to help users reach reflective equilibrium. It functions by filtering out human biases and prejudices while retaining the positive, intuitive aspects of moral reasoning.

Related Experiment Videos

The study aims to introduce the Artificial Moral Advisor as a viable tool for enhancing human moral decision-making. Researchers seek to resolve the tension between objective ethical standards and individual personal values. The project addresses the limitations of human moral psychology by proposing a computational alternative to traditional decision-making processes. The authors intend to demonstrate how a quasi-relativistic version of the ideal observer can function within a digital environment. This work explores the potential for technology to support moral autonomy rather than undermining it. The investigation focuses on overcoming the downsides of human intuition, such as prejudice and bias, while keeping its positive functions. The researchers aim to provide a framework that respects conservative objections regarding the ethics of human bioenhancement. This effort establishes a new direction for integrating philosophical rigor into the development of advanced artificial intelligence systems.

Main Methods:

The authors employ a conceptual analysis approach to bridge classical philosophy and modern computational design. They evaluate the theoretical alignment between the proposed system and Roderick Firth's ideal observer model. The review approach involves identifying specific functional requirements for an impartial, yet non-absolutist, digital agent. Investigators contrast the proposed model with existing frameworks to highlight unique operational characteristics. They synthesize arguments concerning human moral psychology to justify the inclusion of specific decision-making parameters. The study utilizes logical deduction to explore how the system might interact with individual value sets. Researchers examine potential objections to bioenhancement to refine the advisor's operational scope. This methodology focuses on establishing a robust theoretical foundation for future technical implementation.

Main Results:

The primary finding suggests that the Artificial Moral Advisor can successfully integrate impartial judgment with individual moral autonomy. The model demonstrates that a non-absolutist system avoids the rigid constraints of traditional ideal observer frameworks. The authors report that the advisor effectively mitigates human biases while preserving the beneficial functions of emotional intuition. Evidence indicates that the system supports users in achieving both wide and narrow reflective equilibrium. The researchers show that this approach addresses conservative objections to bioenhancement by prioritizing the agent's own principles. The analysis reveals that the advisor remains disinterested and dispassionate, mirroring the consistency of classical philosophical ideals. The study highlights that the technology compensates for inherent limitations in human moral psychology. The findings suggest that this framework provides a structured, reliable method for improving complex ethical deliberations.

Conclusions:

The authors propose that the Artificial Moral Advisor supports individual moral autonomy by integrating personal values into its decision-making framework. This system helps users achieve both wide and narrow reflective equilibrium through consistent, impartial guidance. By addressing specific conservative objections, the model offers a viable path for moral enhancement that avoids common pitfalls. The researchers suggest that the technology effectively captures the positive functions of human intuition while filtering out harmful biases. This synthesis implies that computational agents can serve as valuable partners in ethical deliberation. The authors argue that the advisor maintains a non-absolutist stance, distinguishing it from traditional philosophical ideals. Their analysis indicates that such tools could improve human decision-making by compensating for inherent psychological limitations. The study concludes that this approach provides a structured method for navigating complex moral landscapes without compromising human agency.

The researchers measure success by the system's ability to facilitate both wide and narrow reflective equilibrium. This phenomenon involves aligning personal beliefs with broader ethical principles to achieve a more coherent moral outlook.

The authors claim that this technology respects moral autonomy by enhancing, rather than replacing, human judgment. They propose that the advisor serves as a supportive partner that addresses conservative concerns regarding traditional forms of bioenhancement.