AI bias: exploring discriminatory algorithmic decision-making models and the application of possible machine-centric solutions adapted from the pharmaceutical industry

Abstract

A new and unorthodox approach to deal with discriminatory bias in Artificial Intelligence is needed. As it is explored in detail, the current literature is a dichotomy with studies originating from the contrasting fields of study of either philosophy and sociology or data science and programming. It is suggested that there is a need instead for an integration of both academic approaches, and needs to be machine-centric rather than human-centric applied with a deep understanding of societal and individual prejudices. This article is a novel approach developed into a framework of action: a bias impact assessment to raise awareness of bias and why, a clear set of methodologies as shown in a table comparing with the four stages of pharmaceutical trials, and a summary flowchart. Finally, this study concludes the need for a transnational independent body with enough power to guarantee the implementation of those solutions.

Embedding machine learning based toxicity models within radiotherapy treatment plan optimization

“…A side-advantage of our current approach (learning mathematical representation of ML-based outcome models) is that it adds a level of transparency to the more 'black-box' ML models. This in turn can enhance the users' ability to detect 'nonsensical' decision rules that might have been learned by the models due to, among other factors, model overfitting, insufficient or noisy data, or misrepresentation or bias in the training dataset (Cho 2021, Belenguer 2022, Tasci et al 2022. This is a crucial step, as prediction uncertainty and the lack of physician trust in ML-based outcome models are among the most important hurdles facing the clinical implementation of outcome-based treatment planning.…”

Ethical considerations and concerns in the implementation of AI in pharmacy practice: a cross-sectional study

“…In our context, over half of the participants (56.5%) had concerns about AI systems potentially overselling unnecessary medications and cosmetics to patients, raising questions about beneficence and potential maleficence. Such an act may be the result of a biased algorithmic system in the pharmaceutical market, and the absence of legally binding regulations in such a situation would harm consumers [ 21 ]. AI in pharmacy practice should prioritize patient well-being and ensure that recommendations are based on evidence-based guidelines rather than profit motives.…”

AI Advancements: Comparison of Innovative Techniques

“…These prejudices may show up in automated decision-making procedures, which could result in discrimination in the hiring, lending, and criminal justice systems, among other areas. It is crucial to protect fairness by tackling bias in AI systems to avoid feeding into societal stereotypes [104,105]. Privacy issues are also brought up by the enormous amount of individual data that is necessary for AI to perform at its best [106].…”