Reproduce “Why Machines Can't Be Moral: Turing's Halting Problem and the Moral Limits of Artificial Intelligence” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Massimo Passamonti
Applying Alan Turing's halting problem, this research posits that explicit ethical machines inferring moral principles through bottom-up learning cannot consistently make decisions in all complex moral dilemmas due to computational undecidability, thus fundamentally limiting their capacity for true moral agency.
Key insights
- Moral reasoning, especially when demanding consistency with an AI's learned principles, is computationally intractable for bottom-up explicit ethical machines.
- The halting problem implies that one cannot predict whether such an AI will always halt and produce a decision, or enter an infinite loop, when facing a moral dilemma.
- An AI's potential failure to reliably make a decision due to computational limits, rather than active choice, fundamentally compromises its claim to moral agency.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.