Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Trade-offs in Financial AI: Explainability in a Trilemma with Accuracy and Compliance

Patricia Marcella Evite, Ekaterina Svetlova, Doina Bucur

TL;DR

This paper argues that in finance, explainability cannot be reduced to a simple accuracy–explainability trade-off due to stringent regulatory, latency, and budget constraints. Through twenty qualitative interviews with finance professionals, it introduces the financial XAI trilemma: accuracy and compliance are non-negotiable prerequisites, while ease of understanding determines usable and defensible adoption, with cost and speed acting as operational feasibility levers. The study reframes explainability within a sociotechnical, multi-layered context and highlights the need for compliance-by-design and audience-tailored explanations. Practically, it calls for governance approaches that document trade-offs across stakeholders and use cases to maintain trust, regulatory alignment, and operational viability.

Abstract

As Artificial Intelligence (AI) becomes increasingly embedded in financial decision-making, the opacity of complex models presents significant challenges for professionals and regulators. While the field of Explainable AI (XAI) attempts to bridge this gap, current research often reduces the implementation challenge to a binary trade-off between model accuracy and explainability. This paper argues that such a view is insufficient for the financial domain, where algorithmic choices must navigate a complex sociotechnical web of strict regulatory bounds, budget constraints, and latency requirements. Through semi-structured interviews with twenty finance professionals, ranging from C-suite executives and developers to regulators across multiple regions, this study empirically investigates how practitioners prioritize explainability relative to four competing factors: accuracy, compliance, cost, and speed. Our findings reveal that these priorities are structured not as a simple trade-off, but as a system of distinct prerequisites and constraints. Accuracy and compliance emerge as non-negotiable "hygiene factors": without them, an AI system is viewed as a liability regardless of its transparency. Operational levers (speed and cost) serve as secondary constraints that determine practical feasibility, while ease of understanding functions as a gateway to adoption, shaping whether AI tools are trusted, used, and defensible in practice.

Trade-offs in Financial AI: Explainability in a Trilemma with Accuracy and Compliance

TL;DR

This paper argues that in finance, explainability cannot be reduced to a simple accuracy–explainability trade-off due to stringent regulatory, latency, and budget constraints. Through twenty qualitative interviews with finance professionals, it introduces the financial XAI trilemma: accuracy and compliance are non-negotiable prerequisites, while ease of understanding determines usable and defensible adoption, with cost and speed acting as operational feasibility levers. The study reframes explainability within a sociotechnical, multi-layered context and highlights the need for compliance-by-design and audience-tailored explanations. Practically, it calls for governance approaches that document trade-offs across stakeholders and use cases to maintain trust, regulatory alignment, and operational viability.

Abstract

As Artificial Intelligence (AI) becomes increasingly embedded in financial decision-making, the opacity of complex models presents significant challenges for professionals and regulators. While the field of Explainable AI (XAI) attempts to bridge this gap, current research often reduces the implementation challenge to a binary trade-off between model accuracy and explainability. This paper argues that such a view is insufficient for the financial domain, where algorithmic choices must navigate a complex sociotechnical web of strict regulatory bounds, budget constraints, and latency requirements. Through semi-structured interviews with twenty finance professionals, ranging from C-suite executives and developers to regulators across multiple regions, this study empirically investigates how practitioners prioritize explainability relative to four competing factors: accuracy, compliance, cost, and speed. Our findings reveal that these priorities are structured not as a simple trade-off, but as a system of distinct prerequisites and constraints. Accuracy and compliance emerge as non-negotiable "hygiene factors": without them, an AI system is viewed as a liability regardless of its transparency. Operational levers (speed and cost) serve as secondary constraints that determine practical feasibility, while ease of understanding functions as a gateway to adoption, shaping whether AI tools are trusted, used, and defensible in practice.
Paper Structure (23 sections, 5 figures, 4 tables)

This paper contains 23 sections, 5 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Literature Review: Layered Requirements and Competing Tensions in Financial AI
  3. AI and the accuracy--explainability frontier
  4. Ease of understanding: rethinking explainability for non-technical audiences
  5. A systems-view of AI and demands competing with explainability
  6. Empirical Strategy
  7. Exploratory Interviews
  8. Role and Context
  9. Relevance of Explanation to Work Tasks
  10. Trade-off Ranking Task
  11. Participant Recruitment
  12. Data Analysis
  13. Ethics Statement
  14. AI Deployment Trade-offs as Perceived by Finance Professionals
  15. Accuracy as Prerequisite to Deployment
  16. ...and 8 more sections

Figures (5)

  • Figure 1: The Accuracy-Explainability Frontier in Financial AI
  • Figure 2: Considerations for AI deployment in the financial sector
  • Figure 3: Summary of Interview Participants Demographics
  • Figure 4: Reconstruction of P6’s Alternative Credit Scoring Model
  • Figure 5: The Trilemma in Financial AI