Chatbot apologies: Beyond bullshit

TL;DR

This paper critically examines the nature of apologies made by chatbots, arguing they lack genuine moral agency and are often superficial, raising questions about their design and moral implications.

Contribution

It analyzes the limitations of chatbot apologies by exploring moral and linguistic agency, emphasizing the distinction between superficial and genuine moral expressions.

Findings

Chatbot apologies are superficial and lack moral sincerity.

Genuine apologies require moral and linguistic agency that chatbots do not possess.

Designing chatbots with authentic moral expressions involves complex ethical considerations.

Abstract

Apologies serve essential functions for moral agents such as expressing remorse, taking responsibility, and repairing trust. LLM-based chatbots routinely produce output that has the linguistic form of an apology. However, they do this simply because they are echoing the kinds of things that humans say. Moreover, there are reasons to think that chatbots are not the kind of linguistic or moral agents capable of apology. To put the point bluntly: Chatbot apologies are bullshit. This paper explores this concern and develops it beyond the epithet, drawing on the nature of morally serious apologies, the linguistic agency required to perform them, and the moral agency required for them to matter. We conclude by considering some consequences for how chatbots should be designed and how we ought to think about them.