Clifford B. Anderson

Critical Legal AI Literacies: The Challenges of Synthetic Media

Clifford Anderson speaks on synthetic media as part of the Lillian Goldman Law Library’s Critical Legal AI Literacies series.

Yale Divinity Library Announces New Director

Atla shares the news that Clifford B. Anderson will become Director of the Yale Divinity School Library on May 1, 2024.

Interdisciplinary Team Develops Open-Access Computer Programming Course for High School Teachers

Clifford Anderson, Akos Ledeczi, and Brian Broll launch ‘Programming for a Networked World,’ a beginner-level Coursera MOOC introducing programming, distributed computing, and networking with NetsBlox.

Historians, Data Scientists Work to Preserve Endangered Middle Eastern Culture with NEH Grant

Vanderbilt Divinity School and the Heard Libraries announce a $350,000 NEH grant to make English translations of Syriac literature freely available online.

Automated Speech Recognition Aids in Transcription and Captioning for 62,000 Hours of Archived Television News

The Vanderbilt Television News Archive completes transcription and captioning of its entire collection using Automated Speech Recognition, with a workflow designed by Jim Duran and Cliff Anderson.

Where Does Your Library Stand? A Hands-On AI Maturity Assessment Workshop

AI4LAM Fantastic Futures 2026: Trust in the Loop

This participatory workshop invites attendees to work through the AI Maturity Index for Academic Libraries together and to use their results as a catalyst for structured conversation about AI assessment, planning, prioritization, readiness, and governance gaps. The workshop’s value is not the tool itself; it’s the dialogue the tool generates among library staff with different roles, vantage points, and institutional contexts. The AI Maturity Index for Academic Libraries assesses eight dimensions of library practice—Collections, Discovery, Facilities, Instruction, Metadata, Preservation, Research, and Governance—across five maturity levels. The Index understands that AI maturity is not determined by the amount of AI technology a library uses; instead, it measures how effectively AI serves its users’ needs. Attendees leave with a provisional maturity profile for their institution, facilitation strategies they can replicate with their own staff, and peer connections to colleagues navigating similar challenges.

The Ends of Technology (in Church)

Church & AI Conference / GoNeDiGiTal: Digital Theology to Date and Beyond

My goal in this paper is to explore the crafting of artificial agents that serve the ends of the church. I argue that an existing technique in AI safety training called ‘constitutional AI’ can be fruitfully adapted to create AI agents that promote the glory of God and the flourishing of human beings. After unpacking how constitutional AI works at a technical level—from reinforcement learning with human feedback to reinforcement learning from AI feedback—I trace the evolution of Anthropic’s constitution for Claude from 2023 to 2026 and propose that the church’s long history of confession and catechesis offers a better metaphor than constitutionalism: ‘catechetical AI.’ I then take up questions of free expression, sovereignty, and pluralism, contrasting the subsidiarity of Catholic social teaching in Pope Leo XIV’s Magnifica Humanitas with a Neo-Calvinist model of sphere sovereignty. Bending constitutional AI toward the ends of the church leads inevitably toward a greater pluralism than standard AI safety protocols anticipate—and that, I contend, is a good thing for the future of human–AI collaboration. I forecast the emergence of a new breed of theological engineer who catechizes models as ‘creedal machines’ to serve alongside congregants in the different spheres of human activity.

Slow Minds in Fast Times: Haruki Murakami's Fiction as Metacognitive Resistance to AI

International Conference for the Fantastic in the Arts 47: (Meta)Cognition

This paper (co-authored with Haerin Shin and Douglas Fisher) examines Haruki Murakami’s fiction as a sustained meditation on metacognitive practice that fundamentally opposes the accelerated cognition characteristic of generative artificial intelligence. Murakami rarely writes about computation and evinces distaste for technological advances, yet his fiction offers insight into a fundamental question facing us as AI rapidly advances: how to maintain human ways of knowing in an increasingly automated and algorithmically structured world. Drawing primarily on Hard-Boiled Wonderland and the End of the World, The Wind-Up Bird Chronicle, Dance Dance Dance, Kafka on the Shore, and Killing Commendatore, as well as the nonfiction of Underground, we argue that Murakami’s narrative architectures deliberately cultivate what Daniel Kahneman terms ‘System 2’ thinking—slow, deliberate, consciously effortful cognition. His liminal ‘other worlds’ enforce a durational consciousness that resists algorithmic compression, while his antagonists—hollow figures of algorithmic intelligence like Noboru Wataya and Menshiki—raise the question of whether strategic optimization and pattern recognition, however sophisticated, ever constitute genuine understanding without a moral center or experiential ground. Murakami’s novels model consciousness as essentially inefficient, necessarily embodied, and irreducibly temporal, demonstrating how literature itself might serve as a technology for preserving slow cognition against the tyranny of instantaneous processing.

Generating Stub Articles about Women in Religion: An Experiment in Retrieval Augmented Generation and Fine-Tuning LLMs

Artificial Intelligence and Religion Unit, Annual Meeting of the American Academy of Religion

This paper describes an experiment to generate stub articles about women religious leaders using a purpose-built artificial intelligence system as a means to address gender imbalances on Wikipedia. The Women in Religion User Group is an officially recognized Wikimedia Movement Affiliate that “seeks to create, update, and improve Wikipedia articles pertaining to the lives of cis and transgender women scholars, activists, and practitioners in the world’s religious, spiritual, and wisdom traditions.” (Women in Religion 2025) In the early stages of the project, we explored the use of retrieval augmented generation (RAG) to improve the veracity of the stubs that the LLM generated. In the current phase of the project, we are fine-tuning an open-source large language model to improve its ability to create Wikipedia stubs. After reviewing these techniques, we discuss their effectiveness while also raising ethical questions about releasing our project in open source.

Using AI to Support Editors of Underrepresented Topics on Wikipedia

WikiConference North America 2025

This workshop, facilitated by members of the Women in Religion User Group, explores the ethical, technical, and practical potential of using artificial intelligence to assist editors of underrepresented topics on English-language Wikipedia. We begin by reviewing the Wikimedia Foundation’s April 2025 strategy brief, “Artificial Intelligence for Editors.” Drawing on the experience of our user group, we will discuss ways of “supporting editors with AI” when editing articles about underrepresented areas. We will reflect on our efforts during the past two years to accelerate the production of stubs about notable women in religion using techniques such as retrieval augmented generation, synthetic data creation, and LLM fine-tuning. During our discussion, we will offer a series of pulse surveys to gauge participant sentiment about the employment of these techniques on and off Wiki. As an outcome of this workshop, we will publish our collective thoughts at our user group’s Meta page with the hope of charting a path for using AI responsibly to address knowledge gaps on Wikipedia.

AI as Superscholar: Authorship at the Threshold of the Unsayable

Through a perspectival inversion, the authors explore what artificial intelligence—reconceived as a ‘superscholar’ operating as a scholarly collective unconscious—reveals about limitations in human authorship. AI exposes three interconnected crises: of credit, of verification, of comprehensive knowledge—each compounded by the responsibility gap endemic to learning automata. The article addresses implications for collection development, cataloging, and information literacy, proposing the goal of ’epistemic reparation,’ with AI surfacing marginalized contributions occluded by citational regimes.

Bits and Bias: The Case of Women in Religion

This chapter examines how gender bias becomes encoded in digital knowledge infrastructures, taking the representation of women in religion as a case study.

Kuyper and Princeton

This chapter examines Abraham Kuyper’s relationship with Princeton, from his Stone Lectures at Princeton Theological Seminary in 1898 to Princeton’s later emergence as a center for Kuyper scholarship, tracing the reception of neo-Calvinism in American theological circles.

Artificial Intelligence for Academic Libraries

A comprehensive guide to artificial intelligence for academic library professionals, covering theory, practice, and ethical considerations for integrating AI tools and services into library operations and scholarly support.

The Ethics of Deepfakes for Cultural Heritage: A Hybrid Exhibition of Reanimated Jonathan Edwards as a Test Case

‘Deepfakes,’ synthetic media generated by deep neural networks, have become increasingly common in many sectors, from social media, edtech, and business marketing to museum exhibitions. This trend raises ethical concerns about voice and image manipulation. Unlike living subjects, historical figures cannot contest replications of their image and voice, making ethical considerations crucial. This poster explores ethical issues through our use of an 18th-century historical figure as a test case, documenting strategic and technical challenges. Reflecting on our ethical decisions, we offer recommendations on less-discussed issues such as accent cloning and reanimating figures from diverse historical communities.