Diachronic and synchronic variation in the performance of adaptive machine learning systems: The ethical challenges

Joshua Hatherley, Robert Sparrow · 2025

Read Diachronic and synchronic variation in the performance of adaptive machine learning systems: The ethical challenges by Joshua Hatherley, Robert Sparrow (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2504-08861.txt.

Description

Joshua Hatherley and Robert Sparrow explain the ethical problems raised by medical AI that keeps learning after deployment in a 2025 paper. An adaptive system changes as new patient data arrives, so the tool approved last year may not be the tool in use today. Performance can drift over time at one site, and it can differ at the same moment between sites.

Those two forms of variation, across time and across places, strain consent, safety checks, and blame when harm occurs. The tutorial walks clinicians and regulators through the issues step by step. Health systems adopting self-updating AI get a clear map of what oversight must track when the product never sits still.

Catalog record

CREATORJoshua Hatherley, Robert Sparrow
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE29 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-08861

Browse the full library · About the Tower of Babel