Simon Kolker

A PhD candidate at the University of Manchester

Studying the effect of uncertainty on Machine Ethics. As autonomous systems take more responsibility in the real world, their decisions must reflect our values while handling moral and outcome uncertainty.

Find me on GitHub, Google Scholar, the University of Manchester and LinkedIn.

Projects

See more on my GitHub

Publications

Towards Responsibly Non-Compliant Machines (2026)

Marija Slavkovik, Marie Farrell, Louise Dennis, Michael Fisher, Simon Kolker, Emily C. Collins

In the Rebellious and Disobedient Agents in Artificial Intelligence (RaD-AI) workshop at AAAI Singapore 2026

[Paper Upcoming]

Uncertain Machine Ethics Planning (2025)

Simon Kolker, Louise Dennis, Ramon Fraga Pereira, and Mengwei Xu

24th International Conference on Autonomous Agents and Multiagent Systems

[Paper] [Short Video] [Short Video Slides] [Code]

Applying Ethical Decision Making in Space Missions (2024)

Simon Kolker, Louise Dennis, Ramon Fraga Pereira, and Mengwei Xu

In the New Ideas and Emerging Results Track at SMC-IT/SCC 2024

[IEEE]

Selecting Ethical Actions by Retrospection on Hypothetical Outcomes (2023)

Simon Kolker, Louise Dennis, Ramon Fraga Pereira, and Mengwei Xu

In International Workshop on Computational Machine Ethics (CME) 2023

[Arxiv]

Machine Ethical Decisions with Hypothetical Retrospection (2023)

Simon Kolker, Louise Dennis, Ramon Fraga Pereira, and Mengwei Xu

In International Workshop on Coordination, Organizations, Institutions, Norms and Ethics for Governance of Multi-Agent Systems (COINE) 2023

[Arxiv] [Code] [Springer]

Gallery

dog
In Detroit
Go back!

UTILITARIAN

By remaining a bystander, you have not directly caused harm, though you may have allowed it to occur.

Philippa Foot's famous trolley problem illustrates how our moral intuitions may conflict. Subtle variations of the problem lead many to different judgments.

As we develop machines whose actions have moral consequences, we must consider the kind of morality they should follow.

What if the scenario isn't so simple? If the outcomes are uncertain or the decisions are sequential?