Surface Reading LLMs: Synthetic Text and its Styles

TL;DR

This paper explores how large language models generate synthetic text that mimics human style, emphasizing the importance of surface-level stylistic analysis to understand their impact on meaning-making in society.

Contribution

It introduces a semiotics framework of 'surface integrity' and demonstrates how stylistic markers reveal LLMs as cultural machines shaping discourse.

Findings

Stylistic markers distinguish synthetic from human text.

Surface analysis reveals LLMs' role in transforming meaning circulation.

LLMs act as cultural machines influencing contemporary discourse.

Abstract

Despite a potential plateau in ML advancement, the societal impact of large language models lies not in approaching superintelligence but in generating text surfaces indistinguishable from human writing. While Critical AI Studies provides essential material and socio-technical critique, it risks overlooking how LLMs phenomenologically reshape meaning-making. This paper proposes a semiotics of "surface integrity" as attending to the immediate plane where LLMs inscribe themselves into human communication. I distinguish three knowledge interests in ML research (epistemology, epist\=em\=e, and epistemics) and argue for integrating surface-level stylistic analysis alongside depth-oriented critique. Through two case studies examining stylistic markers of synthetic text, I argue how attending to style as a semiotic phenomenon reveals LLMs as cultural machines that transform the conditions of meaning emergence and circulation in contemporary discourse, independent of questions about machine consciousness.