LLM-impersonated debate contributions are more authentic, relevant and coherent than their original: A representative study using BBC1’s Question Time

Steffen Herbold, Alexander Trautsch, Zlata Kikteva, Annette Hautli-Janisz · 2026

Read LLM-impersonated debate contributions are more authentic, relevant and coherent than their original: A representative study using BBC1’s Question Time by Steffen Herbold, Alexander Trautsch, Zlata Kikteva, Annette Hautli-Janisz (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0347757.txt.

Description

Steffen Herbold and three colleagues published a 2026 study testing whether large language models can convincingly imitate real people in political debate. They drew on Question Time, a BBC television programme where panellists answer audience questions. Then they asked language models to write replies in the panellists' place.

A cross-section of the British public then judged both versions without knowing which was which. The machine-written answers were rated more authentic, more relevant, and more coherent than the originals, and writing style did not explain the gap. The authors warn that such convincing fakes could mislead voters, so the public urgently needs warning about the risk.

Catalog record

CREATORSteffen Herbold, Alexander Trautsch, Zlata Kikteva, Annette Hautli-Janisz
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE48350
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0347757

Browse the full library · About the Tower of Babel