Author or prompter? Scientific writing, identity, and the Theseus paradox.

Scientific authorship is undergoing a subtle but profound transformation. With the rise of generative artificial intelligence (AI), the boundary between human contribution and machine assistance is increasingly blurred. This article explores how AI-assisted scientific writing challenges traditional definitions of authorship, accountability, and intellectual ownership. Drawing on both identity debates and contribution-based accounts, it argues that the growing role of AI in manuscript production demands a reconsideration of contributor roles, transparency, and recognition. Using the Ship of Theseus as a guiding metaphor, the argument shows how authorship can be reconceived when human-authored components are progressively replaced with AI-generated content. The central concern is not whether AI should be credited, but how the human contribution, control, and accountability should be tracked and attributed when AI mediates linguistic or argumentative content. A set of policy measures is proposed-linking existing authorship standards to disclosure thresholds, contributor roles, and an empirical programme for monitoring AI use-to preserve credibility and accountability.