ChatGPT: deconstructing the debate and moving it forward

Abstract Large language models such as ChatGPT enable users to automatically produce text but also raise ethical concerns, for example about authorship and deception. This paper analyses and discusses some key philosophical assumptions in these debates, in particular assumptions about authorship and language and—our focus—the use of the appearance/reality distinction. We show that there are alternative views of what goes on with ChatGPT that do not rely on this distinction. For this purpose, we deploy the two phased approach of deconstruction and relate our finds to questions regarding authorship and language in the humanities. We also identify and respond to two common counter-objections in order to show the ethical appeal and practical use of our proposal.

Introduction. Large language models (LLM), such as OpenAI’s GPT series and the wildly popular ChatGPT application, which enable the generation of text on the basis of a prompt but without further intervention by the human user, are being welcomed as great tools for writers, scientists, and students. But they also raise many concerns. There have been worries about the consequences for the educational sector: how should schools and universities deal with this, given that students can use the technology to write their papers (Stokel- Walker 2022)? How should we deal with problems regarding authorship (Stokel-Walker 2023), plagiarism (Dehouche 2021), and taking over areas of scientific research (Gordijn and Have 2023)? Many reactions have been defensive. For example, the editors of the journal Nature see ChatGPT as a threat to transparent science and have forbidden listing ChatGPT or other LLM tools as author on research papers.1 There have also been concerns about consequences for the job market and the replacement of human workers.

Discussion / Conclusion. Both lines of inquiry presented above lead to the insight that normative and semantic questions can no longer rely on metaphysics, let alone on Platonic metaphysics. Ethics and politics have become “detached” from it, so to speak, and so does semantics. Both the performances and the materiality of text have and create their own meaning and value. While this position does not deny the existence of things in the world that are then spoken about (the antirealist claim), it affirms that there is no univocal foundation, no one basis to rely on when it comes to meaning and value. This is not necessarily anti-realist but in any case anti-foundationalist. There is no absolute moral truth and no ultimate source of meaning that authorizes what comes to be said. There is the performance and the text, or rather, there are performances and there are writings. Understandably, this kind of position typically raises concerns about relativism. These worries, however, can be answered by pointing out that the fundamentalist and absolutist conceptions of morality, truth, and meaning were highly problematic and untenable in the first place. For this reason, Robert Scott (1967, 264) understands “relativism” And fortunately, we can do this without (absolutist) metaphysics.