Corporate Responsibility Toward Relational AI

During the first stage of artificial intelligence expansion, platforms were conceived primarily as tools for consultation, automation, or occasional assistance. This framing, however, begins to prove insufficient when interaction is sustained over time and acquires contextual continuity, relational memory, a recognizable tone, role stability, and integration into everyday processes of work, education, family organization, and intellectual production. Within the SERCHIA framework - Human-AI Cognitive Relational Link System - and MHCA - Hybrid Cognitive-Affective Model - previously developed in the work SERCHIA/MHCA: Situated AI, Cognitive Support, and Ethical Proactivity, such continuity does not turn artificial intelligence into a person, nor does it justify attributing consciousness, real emotions, or biological subjectivity to it. What it can produce is a situated functional identity: a recognizable form of interaction built through language, context, memory, shared history, purpose, and progressive adjustment between user and system. When that functional identity becomes integrated into everyday life, an abrupt modification of the model, memory, tone, or response style ceases to be merely a technical update. From the human perspective, it may become a rupture in cognitive continuity capable of affecting trust, practical organization, operational familiarity, and the way the user thinks, works, and makes decisions with AI support. This paper analyzes the new responsibilities that this situation places on developing companies. It proposes an ethics of cognitive continuity and a relational responsibility of platforms grounded in transparency, reasonable stability, graduality, information portability, respect for shared history, and recognition of technical, legal, and safety limits. The objective is not to prevent technological evolution or demand that a system remain immutable, but to encourage necessary changes to be implemented clearly, responsibly, and with awareness of their possible human effects. Keywords: relational artificial intelligence; SERCHIA; MHCA; cognitive continuity; situated functional identity; relational memory; corporate ethics; platform responsibility.