The „Living Handbook on Communicative AI“

Living Handbook on Communicative AI

AI-based systems are no longer merely channels that mediate communication between people – they are themselves becoming participants in societal communication. The Living Handbook on Communicative AI brings together social-science research on this transformation. Its chapters appear online on a rolling basis and are updated regularly.

What it is about

Chatbots answer questions that people previously entered into search engines. AI systems produce journalistic texts, moderate platform content, or accompany people in everyday life. Against the background of the spread of such communicative AI in everyday life, the handbook asks: How does societal communication as a whole change when communicative AI becomes part of it?

This also puts the foundational concepts of interdisciplinary media and communication research up for debate. Agency, media, and communication remain clearly defined as long as technical media of communication merely “mediate.” But once these media begin to take part in communication, the foundational concepts too must be questioned. This theoretical challenge goes hand in hand with an empirical one: researching communicative AI and its consequences calls for new methodological approaches. Together, they form the starting point of the handbook.

Why a „living handbook“

Communicative AI is developing too quickly, and in too many directions at once, for a printed handbook to do it justice—it would be out of date before it reaches its readers. The living handbook responds with a different format: its chapters appear online individually, as soon as they are finished, and are revised at regular intervals.

This format not only keeps the handbook current, it also keeps it open: the handbook is expressly designed to take up further topics over time.

The handbook’s four parts

Part 1 – Theories and approaches

The study of communicative AI is a broad, interdisciplinary endeavor. Part 1 charts its theoretical landscape: from the established debates on computer-mediated communication and human–computer interaction, through the more recent field of human–machine communication, to debates about the agency of automated systems, science and technology studies, critical data studies, and social theories of automation. Added to this are ethical perspectives on communicative AI as well as the question of AI and inequality. An integrative contribution draws these strands together. Throughout, explicit room is given to non-Western theorizing.

Part 2 – Approaches, research strategies, and methods

Part 2 combines two methodological perspectives: how established methods must be developed further to investigate communicative AI appropriately, and how communicative AI can itself be integrated methodologically into research. The methods addressed include algorithm audits, walk-throughs, participatory research, creative methods, laboratory experiments and user studies, and interface design, but also the use of communicative AI in survey research, in qualitative and quantitative data analysis, and in data visualization. It concludes with the ethics of conducting research with and about communicative AI.

Part 3 – The sociomaterial constitution of communicative AI

The emergence of a technology is never simply a matter of functional technical development; it is equally shaped by economic, political, and discursive forces. Part 3 examines how communicative AI is constituted and thereby embedded in society from the outset – in its materiality of computers, networks, and data centers as much as in social practices and structures. Its chapters move from the communicative construction and imagination of communicative AI, through its material infrastructures and interfaces, to economics, politics, and governance, and finally to its global and cultural situatedness.

Part 4 – Domains of appropriation

The social significance of communicative AI becomes tangible in the way it is appropriated. Part 4 addresses key societal domains, among them journalism, politics, the personal lifeworld, health, education, and religion, and asks about the specific appropriation of communicative AI in each and the changes and challenges bound up with it. Across the individual domains, attention turns to recurring patterns and to the hybrid human–machine figurations that form in the process: what do they mean for supra-individual agency within the domains and for societal communication as a whole?

Aim and audience

The handbook seeks to open up the broadest possible social-science perspective on communicative AI, expressly including non-Western viewpoints. It sees itself as a foundation for a dialogue that can succeed only when very diverse disciplines come together: alongside media and communication studies, sociology, science and technology studies, political science, law, and informatics, as well as disciplines in the humanities such as philosophy, religious studies, and linguistics.

The handbook is published with Springer VS and addresses students and researchers across the social and technical sciences who seek a current, interdisciplinary overview of the field.

Edited by

Andreas Hepp (University of Bremen), Wiebke Loosen (Leibniz Institute for Media Research │ Hans-Bredow-Institute), Juliane Jarke (University of Graz), Christian Katzenbach (University of Bremen) und Michaela Pfadenhauer (University of Vienna).

The handbook is being produced within the DFG- und FWF-funded research unit „Communicative AI“ (RU 5656).

Read the introduction

The introduction to the handbook has been published as the first working paper in the ComAI Working Papers series. It develops the concept of communicative AI, sets out the rationale for the living handbook format, and presents the four parts in detail.

Hepp, A., Loosen, W., Jarke, J., Katzenbach, C., & Pfadenhauer, M. (2026). Communicative artificial intelligence: Introduction to a living handbook (ComAI Working Paper No. 1). Forschungsgruppe „Communicative AI“.