Explainable AI Reloaded: Challenging the XAI Status Quo in the Era of Large Language Models

TL;DR

This paper critiques traditional XAI approaches in the era of Large Language Models, advocating for a human-centered perspective that emphasizes explainability beyond just opening the black-box.

Contribution

It challenges the assumption that opening the black-box is the primary goal of XAI, proposing a shift towards human-centered explainability strategies in LLM contexts.

Findings

Highlights epistemic blind spots of algorithm-centered XAI

Synthesizes XAI research along three dimensions

Proposes a human-centered approach to explainability

Abstract

When the initial vision of Explainable (XAI) was articulated, the most popular framing was to open the (proverbial) "black-box" of AI so that we could understand the inner workings. With the advent of Large Language Models (LLMs), the very ability to open the black-box is increasingly limited especially when it comes to non-AI expert end-users. In this paper, we challenge the assumption of "opening" the black-box in the LLM era and argue for a shift in our XAI expectations. Highlighting the epistemic blind spots of an algorithm-centered XAI view, we argue that a human-centered perspective can be a path forward. We operationalize the argument by synthesizing XAI research along three dimensions: explainability outside the black-box, explainability around the edges of the black box, and explainability that leverages infrastructural seams. We conclude with takeaways that reflexively inform XAI as a domain.