Social Responsibility of Algorithms: An Overview

Abstract

Should we be concerned by the massive use of devices and algorithms which automatically handle an increasing number of everyday activities within our societies? The paper makes a short overview of the scientific investigation around this topic, showing that the development, existence and use of such autonomous artifacts is much older than the recent interest in machine learning monopolised artificial intelligence. We then categorise the impact of using such artifacts to the whole process of data collection, structuring, manipulation as well as in recommendation and decision making. The suggested framework allows to identify a number of challenges for the whole community of decision analysts, both researchers and practitioners.

Deep Automation Bias: How to Tackle a Wicked Problem of AI?

“…With AI becoming more widespread, there is an intensified debate on issues regarding algorithmic bias, discrimination and debiasing by fostering FAT in ML. Many scholars from various disciplines have been pointing out that algorithms and particularly ML are biased, unfair and lack in transparency and accountability ( [1][2][3][4][5][6][7][8][9][10][11][12]). To deal with these serious issues, there is important research proceeding in the AI/ML community, particularly in computer science, data science and related disciplines aiming at developing approaches for debiasing and improving FAT ( [3,4,[12][13][14][15]).…”

“…While it is obviously relevant to tackle these issues of data quality and technical bias, there is also a need for approaches to address the societal and ethical issues of AI. Several scholars thus argue that bias is not merely a technical issue ( [1,[3][4][5][6][11][12][13][14][15][16][17]). The complexity of the problem is already observable in the various different, and partially contradictory notions and definitions of fairness ( [5,11]).…”

“…Several scholars thus argue that bias is not merely a technical issue ( [1,[3][4][5][6][11][12][13][14][15][16][17]). The complexity of the problem is already observable in the various different, and partially contradictory notions and definitions of fairness ( [5,11]). Similar is the case for the other concepts of FAT.…”

“…There is no general definition for concepts such as fairness, unbiased, neutral, etc. [11]. Transparency, which is basically essential, and which is not just an ethical principle but an enabling condition for ethical practices ( [5,17]), can also have ambiguous meanings.…”

“…To deal with such risks, it is important to recall that automation is not an inevitable path we must take with AI; or as [11] points out (p. 5): "automation is not a straightforward perspective, but a choice." Consequently, "there is somebody who makes the choice and there should be reasons" justifying this choice.…”