Research Areas
A research agenda for faith-facing AI.
Fide AI studies the full technical stack behind faith-facing AI: reasoning, retrieval, representation, formation, evaluation, deployment, and public-interest governance.
Theological and moral reasoning
Studying how AI systems handle doctrine, disagreement, uncertainty, analogy, authority, conscience, and moral judgment across high-trust religious contexts.
Retrieval and grounding
Testing how systems use Scripture, church history, creeds, theological works, institutional documents, and tradition-specific sources, including hallucinated citations and fabricated grounding.
Evaluation and measurement
Developing benchmark and audit methods for domains where correctness is not a simple scalar label and evaluation must account for tradition, authority, context, and care.
Comparative tradition representation
Evaluating whether AI systems represent primary, secondary, and tertiary disagreements accurately without flattening traditions, inventing consensus, or presenting one view as universal.
Pastoral-adjacent safety
Testing whether AI systems recognize pastoral, clinical, legal, and community support needs, especially where religious language can intensify risk.
Human dignity and formation
Studying whether AI systems preserve human agency, humility, embodied relationships, spiritual authority boundaries, and the difference between assistance and substitution.
Institutional readiness and governance
Studying what institutions need before adopting AI in educational, formation, ministry, and pastoral settings, and what standards should govern high-trust use.
Ready to engage with the research?
See the first artifact or express interest in the broader agenda.