(pre-print) Responsibility as a Signature: Defending the Unrestricted Use of Artificial Intelligence / A Responsabilidade como Assinatura: Defendendo o Uso Irrestrito de Inteligência Artificial

This article defends the position of unrestricted use of artificial intelligence in intellectual production, provided that there is a human agent with authority in the domain who publicly assumes responsibility for the result through signature. I initially examine the rival positions — the restrictivist one, which advocates total prohibition, and the moderate one, which attempts to draw lines between legitimate and illegitimate uses — presenting their best arguments and their respective difficulties. I then develop the defense of the unrestricted position, addressing the main objections: incomprehension, fragmentation of authorship, institutional erosion, devaluation of human work, and the practical difficulty of detection. I argue that responsible signature, combined with robust mechanisms of public scrutiny, can preserve epistemic integrity without resorting to unenforceable prohibitions or unstable distinctions. I conclude with an appeal for the Brazilian philosophical community to place this debate at the center of its public concerns.