In this paper, I offer a reply to Lavazza (2025), who states that using an LLM as a co-author to apologize to a partner – far from compromising second-person authen- ticity – can improve relational morality. I agree that some valuable aspects of the relationship may be enhanced in the co-authorship context; however, the second- person authenticity standard cannot still be adequately met. At best, using ChatGPT as a co-author might be compatible with a scenario in which the second-person authenticity assessment is suspended and postponed to a later point in time; this allows us to capture a potential diachronic dimension in such an assessment that has not yet been explored. In the remainder of the article, I criticize Lavazza’s Kantian argument and offer some clarifications on the analogy between using ChatGPT and friend-suggested communication, as well as on the terminology used.
Co-authoring Apologies with ChatGPT Can Support Relationships but not Second-Person Authenticity: A Reply to Lavazza
Battisti, Davide
2025
Abstract
In this paper, I offer a reply to Lavazza (2025), who states that using an LLM as a co-author to apologize to a partner – far from compromising second-person authen- ticity – can improve relational morality. I agree that some valuable aspects of the relationship may be enhanced in the co-authorship context; however, the second- person authenticity standard cannot still be adequately met. At best, using ChatGPT as a co-author might be compatible with a scenario in which the second-person authenticity assessment is suspended and postponed to a later point in time; this allows us to capture a potential diachronic dimension in such an assessment that has not yet been explored. In the remainder of the article, I criticize Lavazza’s Kantian argument and offer some clarifications on the analogy between using ChatGPT and friend-suggested communication, as well as on the terminology used.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
