AI Anthropology is an emerging field that examines artificial intelligence systems through anthropological methods and, simultaneously, uses AI to extend anthropological research and practice. Unlike approaches that position anthropology solely as a critic of technology, AI Anthropology treats AI as a domain where cultural and computational processes are inseparable, and where anthropologists can contribute not only analysis but design, infrastructure, and tools.
The field operates across three interconnected dimensions. Anthropology of AI examines AI systems as culturally situated artifacts, studying how they are designed, deployed, experienced, and contested across social contexts. Anthropology by AI extends anthropological inquiry with AI-enabled computational methods, using tools like large language models, knowledge graphs, and agent-based systems to support ethnographic research at scales and speeds that manual methods alone cannot achieve. Anthropology for AI embeds anthropological insight into AI design, contributing cultural reasoning, contextual judgment, and ethical grounding to the systems being built. These three dimensions are not sequential stages but ongoing, iterative relationships, organized through the AI Anthropology Lifecycle (AAL).
Why AI Anthropology matters now
Anthropologists have studied technology for decades, and computational approaches to social science are not new. What makes AI Anthropology distinct is the current moment: large language models, recommender systems, and agentic AI tools are reshaping how knowledge is produced, circulated, and consumed. These systems are not neutral infrastructure. They encode values, reproduce biases, and create new forms of cultural practice. At the same time, they offer anthropologists tools that can extend the discipline’s reach in ways that were not possible five years ago.
The challenge anthropology faces is that these systems are being built with or without anthropological input. If anthropologists engage only as critics after the fact, the discipline risks repeating the pattern of social media: abundant critique, too-late engagement, and development ceded to others. AI Anthropology argues that anthropologists should be building, not just studying, and that the discipline has the theoretical and methodological resources to do so.
Key concepts and methods
AI Anthropology draws on and has produced several original concepts and methods:
- Multi-Agent Ethnography (MAE) positions AI agents as configurable collaborators within human-AI research networks, extending anthropology’s tradition of multi-sited and multi-sensory approaches.
- Automated Digital Ethnography (ADE) deploys AI agents to continuously collect and analyze data within digital field sites in real time.
- Epistemological Prompting structures prompts to configure AI agents for specific analytical commitments, whether positivist, interpretivist, critical, or pragmatic.
- Machine Knowing describes the shift from statistical pattern-matching to knowledge-grounded cultural interpretation, achieved by combining LLMs with an Anthropological Knowledge Graph.
- Co-Becoming provides a theoretical framework for the mutual transformation that occurs when humans and AI systems work together over time.
- Behavioral Capital offers a metric for evaluating user participation in digital platforms based on engagement and cooperation rather than social or economic status.
These concepts and methods are operationalized in the AI Anthropology Toolkit, an open-source set of computational tools for qualitative research released on GitHub and Zenodo.
Institutional context
AI Anthropology is being developed at the intersection of applied anthropology, science and technology studies, and computational social science. Matt Artz introduced the call for AI Anthropology as a distinct field in General Anthropology (2026) and develops the concept through his PhD in Computational Anthropology at Aalborg University’s Techno-Anthropology Lab, under the supervision of Torben Elgaard Jensen. His dissertation is titled Toward an AI Anthropology.
The work builds on a tradition of combining ethnographic and computational approaches developed at TANTLab by Jensen, Anders Kristian Munk, and Anders Koed Madsen, and it draws on broader conversations in digital ethnography, design research, and the anthropology of technology.
Publications
Key publications in AI Anthropology include:
- “A Call for an AI Anthropology.” General Anthropology 33(1), 2026.
- “Multi-Agent Ethnography: Evolving Anthropological Practice Through Human-AI Collaboration.” Anthropological Forum, 2026.
- “From Machine Learning to Machine Knowing: A Digital Anthropology Approach for the Machine Interpretation of Cultures.” UNESCO Digital Library, 2023.
- “Grounding Large Language Models in an Anthropological Knowledge Graph: A Neuro-Symbolic Approach.” Human Organization, forthcoming.
- “Ten Predictions for AI and the Future of Anthropology.” Anthropology News 64(2), 2023.
- “The Digital Turn in Business Anthropology.” Journal of Business Anthropology 12(1), 2023.