Societal Negotiation of the Status of Intelligence

Dickel examines the conditions under which machines are recognized as “intelligent” within contemporary society. His core premise is that artificial intelligence is not a clearly delimited technical object, but rather a culturally contested category that is continuously reconstituted through successive waves of technological development. Whether a technology is perceived and acknowledged as AI thus depends less on its internal functional mechanisms than on the expectations projected onto it—and on the ways in which it manifests itself within, and intervenes in, everyday social life.

The Shifting Boundary Between Humans and Machines

Central to Dickel’s analysis is the observation that contemporary AI systems increasingly challenge established distinctions between humans and machines. Whereas earlier forms of AI primarily operated in the background of technical infrastructures, current systems appear as visible and audible communicative counterparts that may assume a significant role in social interactions. They diffuse from narrowly defined expert domains into virtually all spheres of society. This transformation gives rise to an intermediate zone in which machines are no longer merely instrumental tools, but increasingly appear as quasi-autonomous actors, collaborators, cognitive partners, and co-decision-makers. Consequently, previously stable boundary constructions—particularly those that normatively distinguish human intelligence as naturally given from machine intelligence—are destabilized.

Communicative AI: The Centrality of Interfaces

One of Dickel’s key arguments is that machines are perceived as intelligent insofar as they appear capable of communication. The decisive concept here is that of communicativization. What matters is not primarily what machines are technically able to do, but how they present themselves to users and how they affect them. This is where specifically designed interfaces become crucial: chat windows, voice-based dialogues, and visual interfaces create mediated zones of interaction in which machines can imitate human modes of expression. While complex algorithmic processes remain opaque, users encounter only the interface—producing a form of “communicative resemblance” that situates machines in social proximity and renders them plausible as conversational partners.

The AGI Discourse as a Future-Oriented Narrative

Dickel also addresses the renewed debate surrounding Artificial General Intelligence (AGI). Rather than constituting a realistically attainable short-term technological objective, AGI functions primarily as a future-oriented narrative that bridges uncertainty regarding the actual status of present-day AI systems. This narrative keeps the notion of “genuine,” human-like intelligence open as a future possibility and thereby establishes a horizon of expectations within which current AI systems can be interpreted as preliminary stages. In this way, the AGI discourse symbolically stabilizes what remains technologically underdetermined: the ontological status of AI.

AI Between Fiction and Reality

The Mainz-based sociologist concludes that AI currently occupies an ontologically hybrid position: it is simultaneously a material technical system and a cultural projection. The societal efficacy of AI derives not only from its algorithmic capacities, but also from the manner in which it communicates via interfaces, consolidates expectations, and challenges established social classifications. Precisely because AI oscillates between fiction and reality, its societal status is continuously subject to renegotiation. According to Dickel, this dynamic constitutes a central driver of the contemporary AI boom.

The podcast discussion does not neglect the societal risks associated with these developments, including questions of power, responsibility, and ethics. At the same time, the guest acknowledges the practical advantages of AI applications and reports making frequent and diverse use of language models in everyday contexts—an experience that, like that of many users, alternates between highly useful and largely unproductive outcomes.

The episode draws on key arguments from Sascha Dickel’s work and develops further reflections in relation to current debates on generative AI and the use of large language models. The conversation is conducted by Prof. Dr. Marcel Schütz and research associate Philipp Krüger from the KIWIT research group.

Reference
Dickel, Sascha (2025): Im Imitationsspiel. Über die Kommunikation mit Maschinen und das Streben nach Artificial General Intelligence. Zeitschrift für Soziologie, 54(2), 190–206. https://doi.org/10.1515/zfsoz-2025-2011