Workshop on Formal Methods for Machine Ethics and Emergent Machine Ethics

Overview

Machine ethics is concerned with the behaviour of machines towards people and other machines. Advances in Artificial Intelligence (AI) continue to reinforce the need for research work in this field either in the top down, bottom up or hybrid approaches. In one direction this work involves specifying, implementing, and verifying ethical and safe AI. In another direction is the study of ethics that emerges by self-organizing inside AI-centric societies. We need to be able to verify the existing ethical capabilities of AI systems, and make advances towards understanding how transparent, collaboratively guidable AI ecosystems might evolve. All of these efforts, while happening at different technology and conceptual levels, are united in the need to establish a global hub for AI-safety, multi-agent cooperation, and governance research.

Twenty years ago, a fall AAAI Fall symposium on Machine Ethics, kick-started the machine ethics field in computer science (https://auld.aaai.org/Library/Symposia/Fall/fs05-06.php ). The time is right for a formative venue of a new generation of machine ethics researchers and teams. Building on emerging work in communities such as Japan's SIG-AGI ( https://www.sig-agi.org/sig-agi/event/sig-agi-30-panel-en ), where foundations for Emergent Machine Ethics are being actively developed, this workshop marks a critical evolution from traditional top-down approaches to include bottom-up emergence, recognizing that as AI systems become increasingly autonomous, we need both formal verification methods and emergent approaches working in tandem to respect their agency while ensuring beneficial outcomes.This workshop has two objectives.

Consolidation: investigation of formal methods for and in machine ethics and value alignment.

We aim to provide an arena for presenting new work on classical topics studied in machine ethics from an artificial intelligence and computer science perspective. We hope to see and discuss to tackling challenges of how autonomous and intelligent systems

systems can acquire verifiably correct behavior: how we can specify, implement and validate ethical and safe reasoning.

Emergence: investigation of Emergent Machine Ethics (EME) - ethics that self-organize within AI-centric societies. As AI capabilities advance exponentially, the rapid pace of development, dynamic human values, and emergent AI-AI interactions necessitate ethics that develop internally within AI systems to complement traditional approaches. EME encompasses studying ethical emergence dynamics, developing inter-intelligence evaluation systems, and facilitating human-AI co-creative guidance. Through this paradigm, we seek to understand how AI societies might develop internal ethical capabilities that scale with their increasing autonomy and complexity.

Important Dates

EventDate (AoE) ??

• Submission Deadline: November 10, 2025

• Notification of Acceptance: November 25, 2025

• Camera‑ready & schedule online xxxx

• Workshop day January 26, 2026