Escrivá establishes that technology is one of the "immense panoramas of work" in which God awaits us — the theological ground for Principle I (Canonical Fidelity) and the conviction that building SPI is itself an act of sanctification.

Del Portillo grounds the distinction between mediation and sacramental reality — the canonical basis for Principle III (Sacramental Distinction): SPI accompanies, never replaces, what belongs to the ordained ministry.

Echevarría identifies the digital world as one of the "new areopagoi" — the pastoral mandate behind Principle IV (Pastoral Oversight) and the formation program's focus on young people forming their faith in digital environments.

The foundational paper that introduced the Transformer architecture — the technical basis of every large language model in existence today, including the AI systems used by SPI. Google Brain / Google Research. Advances in Neural Information Processing Systems, vol. 30.

Why it matters: Understanding that LLMs are "next-token predictors" trained on statistical patterns — not reasoning systems with comprehension — is the basis of SPI's transparency principle and Module I of the formation program.

The GPT-3 paper (OpenAI) that demonstrated large-scale language models could perform complex tasks with minimal instruction. The moment when AI conversational systems became publicly visible. Advances in Neural Information Processing Systems, vol. 33.

Why it matters: GPT-3 marked the beginning of the widespread phenomenon of young people forming "relationships" with AI chatbots — the pastoral problem that Module III of SPI's formation program directly addresses.

Anthropic's research on training AI systems to be helpful, harmless, and honest — including their "Constitutional AI" approach: training models using a written set of principles rather than purely human feedback. This is the system behind the AI guide within SPI's VR experience.

Why it matters: SPI chose Claude (Anthropic) specifically because its design philosophy is closest to the principle that values must be embedded in the system, not added later — which is precisely what Magnifica Humanitas §104 requires.

The paper that introduced the RAG (Retrieval-Augmented Generation) architecture: instead of relying on a model's internal training data, the system retrieves relevant documents from a curated database and uses them to ground responses. Facebook AI Research (Meta). NeurIPS 2020.

Why it matters: SPI's AI guide does not "know" the Gospels or the Catechism from general training — it retrieves passages from a controlled theological corpus. This is what makes doctrinal fidelity technically enforceable, not merely intended.

The most rigorous treatment of the AI alignment problem — what it means for an AI system to pursue objectives that are actually good for humanity — by one of the founding figures of AI research. Viking / Penguin Random House.

Why it matters: Russell's argument that AI systems must be designed with explicit uncertainty about human values — and must defer to human judgment rather than optimize for a fixed objective — is the secular equivalent of what León XIV calls "desarmar la IA" in Magnifica Humanitas §110.

Director of the Stanford Virtual Human Interaction Lab (VHIL). Twenty years of empirical research on VR's effects on empathy, behavior change, and psychological presence. W. W. Norton & Company.

Why it matters: Provides the scientific evidence base for why VR can deepen formation — and why it requires ethical guardrails. The same mechanisms that generate empathy can be misused. SPI's Principle V (Human Dignity and Equity) responds to this research directly.

Foundational research on "presence" — the subjective sense of "being there" in a virtual environment — and its neurological and behavioral correlates. Mel Slater (University of Barcelona) is one of the defining researchers in VR presence science. Frontiers in Robotics and AI.

Why it matters: The concept of presence explains why the Sermon on the Mount in VR is qualitatively different from watching a film of it. This distinction is the foundation of SPI's claim that VR can serve as a form of immersive contemplation.

Introduced the "Reality–Virtuality Continuum" — the conceptual framework that defines the spectrum from fully real to fully virtual environments, including Mixed Reality and Augmented Reality. IEICE Transactions on Information and Systems.

Why it matters: Provides the technical vocabulary for distinguishing what kind of experience SPI offers — and why full immersion (not AR overlay) was chosen for biblical encounters.

The open web standard that enables VR and AR experiences in a web browser without app store installation. A-Frame is Mozilla's open-source framework built on WebXR and Three.js. SPI's current prototype runs entirely on this stack — accessible on Meta Quest, mobile, and desktop without downloading anything.

Why it matters: The choice of WebXR was a deliberate ethical decision: no paywall, no app store gatekeeping, no proprietary platform. An experience of the Sermon on the Mount should be reachable from any smartphone — consistent with Magnifica Humanitas §108–110 on equitable access.

Foundational academic text on how religious communities adapt and engage with digital technologies. Heidi Campbell (Texas A&M) coined the "networked religion" concept. Routledge.

Why it matters: Establishes the scholarly framework for understanding why digital religious initiatives require both theological legitimacy and sociological awareness — not just technical capability.

MIT sociologist's analysis of how digital relationships — including with AI — erode authentic human connection by offering simulation at the cost of genuine encounter. Basic Books.

Why it matters: Turkle's research is the empirical foundation of the pastoral problem that SPI's formation program addresses: young people are not choosing AI because they prefer it — they turn to it because no one has given them a better framework, as León XIV observed in Magnifica Humanitas §100.

The Vatican's first document specifically addressing the relationship between human intelligence and artificial intelligence. Affirms that human intelligence — created in the image of God (imago Dei) — cannot be replicated by machines, while acknowledging AI as a tool for human flourishing when properly governed.

Why it matters: Provides the doctrinal framework — more specific than Magnifica Humanitas — for how Catholic institutions should understand and deploy AI. SPI's Principle II (Ontological Transparency) directly implements Antiqua et Nova's core requirement.

Comprehensive treatment of AI ethics, including value alignment, autonomy, and AI deployment in sensitive human contexts. Oxford University Press.

Why it matters: Floridi's framework for "information ethics" maps directly onto SPI's requirement that the user always knows they are interacting with a mediated representation — never with a person or with the divine.