Artificial Intelligence Transparency Under the European Union AI Act: A Cross-Sectional Audit of Museum-Facing Chatbots

Since August 2026, Article 50 of the EU AI Act has required that people be informed, at the latest at their first interaction, that they are dealing with an AI system; Commission guidelines add disclosure whenever asked. We examine how this duty travels through providers and deployers to the interface visitors see. We conducted a pre-registered cross-sectional audit of discoverable museum-facing chatbots in the EU-27. Of 142 systems identified, 104 in 16 Member States were probed with a fixed protocol, including an identity question asked in two sessions, and coded against two operational disclosure criteria. In total, 76.9% disclosed their AI nature at or before the first interaction and 77.9% when asked in both sessions. Disclosure at launch was 30 percentage points lower for museum-run systems than for municipal or tourism portal systems (61.0% vs. 91.2%), a gap that persisted, attenuated, across platform and country checks. The criteria failed in different systems: 20 answered correctly but showed no notice; 19 showed a notice but did not disclose when asked. Observable disclosure differed sharply by supplier type, suggesting product defaults as a lever for guidance, procurement and enforcement.

Authors

Publication Details

Journal
Preprints.org
Published
2026-11-18
DOI
https://doi.org/10.20944/preprints202611.1830.v1
Primary Topic
Digitalization, Law, and Regulation
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Artificial Intelligence Transparency Under the European Union AI Act: A Cross-Sectional Audit of Museum-Facing Chatbots

Markos Konstantakis
Preprints.org
Digitalization, Law, and Regulation
preprint

Artificial Intelligence Transparency Under the European Union AI Act: A Cross-Sectional Audit of Museum-Facing Chatbots

Markos Konstantakis
preprint en

Abstract

Since August 2026, Article 50 of the EU AI Act has required that people be informed, at the latest at their first interaction, that they are dealing with an AI system; Commission guidelines add disclosure whenever asked. We examine how this duty travels through providers and deployers to the interface visitors see. We conducted a pre-registered cross-sectional audit of discoverable museum-facing chatbots in the EU-27. Of 142 systems identified, 104 in 16 Member States were probed with a fixed protocol, including an identity question asked in two sessions, and coded against two operational disclosure criteria. In total, 76.9% disclosed their AI nature at or before the first interaction and 77.9% when asked in both sessions. Disclosure at launch was 30 percentage points lower for museum-run systems than for municipal or tourism portal systems (61.0% vs. 91.2%), a gap that persisted, attenuated, across platform and country checks. The criteria failed in different systems: 20 answered correctly but showed no notice; 19 showed a notice but did not disclose when asked. Observable disclosure differed sharply by supplier type, suggesting product defaults as a lever for guidance, procurement and enforcement.

Preprints.org
Digitalization, Law, and Regulation
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon