AI-enabled systems are increasingly deployed across various domains, many of which function as 'black boxes.' This means users lack visibility into how inputs lead to outputs.
The combination of autonomy and opacity raises concerns regarding system reliability, consistency, human control, and compliance with ethical and legal norms.
This article argues that many of these worries can be mitigated by leveraging practical judgment, virtue, and intuition when deploying and using opaque AI-enabled systems. We demonstrate that focusing on these distinctly human capabilities creates a bridge between the practical challenges posed by opacity and the ethical, legal, and social norms that govern specific domains.
A key element is recognizing that many positive human traits are unquantifiable, necessitating the development of training regimens and guidelines for AI deployment anchored in humanistic but non-quantifiable values.
Throughout the article, we focus on the military domain as an exemplar of the importance of practical judgment, virtue, and intuition in ethical and effective decision-making regarding AI deployments, though the underlying arguments apply to all domains where opaque and potentially autonomous systems are deployed (with domain-specific modifications).
Blogger's Review: This article emphasizes how moral judgment and intuition can ensure the ethical use of increasingly opaque AI systems. Particularly in military contexts, this exploration of methodology provides a crucial ethical perspective for future AI applications, which is worth considering across various industries.