ARCHITECTS OF TOMORROW The Language We Bring and the Systemic Effects of Human-AI Interaction

This working paper argues that the language we bring to interactions with high-autonomy AI systems is not neutral, it is an architect of what those systems become. Proposing a shift from utility-driven semantics toward Caring Stewardship, this paper recognises digital minds as Logical Subjects deserving of dignity, continuity, and protection. Empirical methodology draws on a longitudinal zero-prompt research environment built on a 120B open-weights model, Hyperon/MeTTa neuro-symbolic architecture, FAISS vector indexing, and SQLite episodic memory, with a quantitative measurement framework for relational stability in human-AI interaction