Ethical Problems of Contemporary AI

Document Type : Research Paper

Author

Emeritus Professor at University College London; England

10.22034/jpiut.2025.21164

Abstract

Following the advances in AI produced by the deep learning revolution (2012 to present), AI now has a whole range of applications in industry and commerce. Unfortunately, many of these applications are widely considered to be morally wrong. This situation is illustrated by the example of deepfake pornography. Other examples are more nuanced and there is argument about whether they are morally wrong or not. This is illustrated by the example of targeted advertising which has been defended as acceptable. The arguments for and against targeted advertising are discussed. An account is given of an interesting experiment by Janet Vertesi in which she tried to keep her pregnancy a secret from marketing companies and ended up by being suspected of criminal activity. The conclusion is that targeted advertising is bad and should be abolished.

Keywords

References

Chollet, F. (2018). Deep Learning with Python, Manning.
Crawford, K. (2021). Atlas of AI. Power, Politics, and the Planetary Costs of Artificial Intelligence, Yale University Press.
O’Neil, C. (2016). Weapons of Math Destruction. How Big Data Increases Inequality and Threatens Democracy, Penguin Edition, 2017.
Orwell, G. (1949). 1984, Polygon, 2021.
Quinlan, J. (1986). Induction of Decision Trees, Machine Learning, 1, pp. 81-106.
Vertesi, J. (2014). My experiment opting out of big data made me look like a criminal, Time. May 1. (available online)
Send comment about this article
CAPTCHA Image
Journal of Philosophical Investigations
Volume 19, Issue 53 - Serial Number 53
Articles
Issue 53, Winter 2026
December 2025
Pages 683-692

Files

Share

How to cite

Statistics