Speculations on Uncertainty and Humane Algorithms
Nicholas Gray
TL;DR
The paper addresses the problem that AI systems in high-risk domains can cause harm when uncertainty and provenance are overlooked. It argues for embedding both aleatory and epistemic uncertainty into AI design, and for robust provenance tracking to bolster trust and guard against misinformation and hallucinations. The approach emphasizes communicating uncertainty in intuitive formats (e.g., intervals, natural frequencies) and ensuring humane human–algorithm interaction rather than relying solely on empirical accuracy metrics. The work aims to provide a conceptual framework and design guidance for building trustworthy, transparent, and humane AI that performs safely in high-stakes settings.
Abstract
The appreciation and utilisation of risk and uncertainty can play a key role in helping to solve some of the many ethical issues that are posed by AI. Understanding the uncertainties can allow algorithms to make better decisions by providing interrogatable avenues to check the correctness of outputs. Allowing algorithms to deal with variability and ambiguity with their inputs means they do not need to force people into uncomfortable classifications. Provenance enables algorithms to know what they know preventing possible harms. Additionally, uncertainty about provenance highlights the trustworthiness of algorithms. It is essential to compute with what we know rather than make assumptions that may be unjustified or untenable. This paper provides a perspective on the need for the importance of risk and uncertainty in the development of ethical AI, especially in high-risk scenarios. It argues that the handling of uncertainty, especially epistemic uncertainty, is critical to ensuring that algorithms do not cause harm and are trustworthy and ensure that the decisions that they make are humane.