A Misdirected Principle with a Catch: Explicability for AI

Abstract

There is widespread agreement that there should be a principle requiring that artificial intelligence (AI) be 'explicable'. Microsoft, Google, the World Economic Forum, the draft AI ethics guidelines for the EU commission, etc. all include a principle for AI that falls under the umbrella of 'explicability'. Roughly, the principle states that "for AI to promote and not constrain human autonomy, our 'decision about who should decide' must be informed by knowledge of how AI would act instead of us" (Floridi et al. in Minds Mach 28(4):689-707, 2018). There is a strong intuition that if an algorithm decides, for example, whether to give someone a loan, then that algorithm should be explicable. I argue here, however, that such a principle is misdirected. The property of requiring explicability should attach to a particular action or decision rather than the entity making that decision. It is the context and the potential harm resulting from decisions that drive the moral need for explicability-not the process by which decisions are reached. Related to this is the fact that AI is used for many low-risk purposes for which it would be unnecessary to require that it be explicable. A principle requiring explicability would prevent us from reaping the benefits of AI used in these situations. Finally, the explanations given by explicable AI are only fruitful if we already know which considerations are acceptable for the decision at hand. If we already have these considerations, then there is no need to use contemporary AI algorithms because standard automation would be available. In other words, a principle of explicability for AI makes the use of AI redundant.

What do we want from Explainable Artificial Intelligence (XAI)? – A stakeholder perspective on XAI and a conceptual model guiding interdisciplinary XAI research

“…Controllability Retain (complete) human control concerning a system User [6,12,13,22,44,49,56,60,74] [39]…”

“…Privacy Assess and increase a system's privacy practices User [14,16,78,98] Responsibility Provide appropriate means to let humans remain responsible or to increase perceived responsibility Regulator [6,13,20,43,56,57,60,104] [68]…”

“…Gain scientific insights from the system User [5,7,12,14,35,37,44,47,60,63,66,67,70,87,97,100,106] Security Assess and increase a system's security [13,16,25,26,29,33,34,37,39,44,46,49,55,57,62,63,67,74,75,78,80,92,94,107,108] [ 28,31,43,50] Trust Calibrate appropriate trust in the system User, Deployer…”

What does it mean to embed ethics in data science? An integrative approach based on microethics and virtues

“…This emphasizes another aspect of digital microvirtue ethics, which is the ability to identify the relevant stakeholders and not just focusing on the immediate recipients of the machine learning system. This also implies that, sometimes, we identify relevant stakeholders but we think that there are no particular moral concerns attached to them (Robbins 2019). With this ability, we direct moral attention and concern to those we think deserve it.…”

Sustainable AI: AI for sustainability and the sustainability of AI

“…Superintelligence) and amounted to the ethics of fanciful scenarios of robot uprisings [6]. The second wave of AI ethics addressed the practical concerns of machine learning (ML) techniques: the black-box algorithm and the problem of explainability [9,16], the lack of equal representation in training data and the resulting biases in AI models [2,7], and the increase in facial and emotion recognition systems infringing on citizen's rights (e.g. privacy) [4].…”