Emerging technology is slowly finding a place in developing countries for its potential to plug gaps in ailing public service systems, such as healthcare. At the same time, cases of bias and discrimination that overlap with the complexity of algorithms have created a trust problem with technology. Promoting transparency in algorithmic decision-making through explainability can be pivotal in addressing the lack of trust with medical artificial intelligence (AI), but this comes with challenges for providers and regulators. In generating explainability, AI providers need to prioritise their accountability to patient safety given that the most accurate of algorithms are still opaque. There are also additional costs involved. Regulators looking to facilitate the entry of innovation while prioritising patient safety will need to look into ascertaining a reasonable level of explainability considering risk factors and the context of its use, and adaptive and experimental means of regulation. Artificial intelligence (AI) models across the globe have come under the scanner over ethical issues; for instance, Amazon's hiring algorithm reportedly discriminates against women,[1] and there is evidence of racial bias in the facial recognition software used by law enforcement in the United States (US).[2] While biased AI has various implications, concerns around the use of AI in ethically sensitive industries, such as healthcare, justifiably require closer examination. Medical AI models have become more commonplace in clinical and healthcare settings due to their higher accuracy and lower turnaround time and cost in comparison to non-AI techniques.

Abstract: The recent enthusiasm for artificial intelligence (AI) is due principally to advances in deep learning. Deep learning methods are remarkably accurate, but also opaque, which limits their potential use in safety-critical applications. To achieve trust and accountability, designers and operators of machine learning algorithms must be able to explain the inner workings, the results and the causes of failures of algorithms to users, regulators, and citizens. The originality of this paper is to combine technical, legal and economic aspects of explainability to develop a framework for defining the "right" level of explain-ability in a given context. We propose three logical steps: First, define the main contextual factors, such as who the audience of the explanation is, the operational context, the level of harm that the system could cause, and the legal/regulatory framework.