Transparent AI Disclosure Obligations: Who, What, When, Where, Why, How

Abdallah El Ali; Karthikeya Puttur Venkatraj; Sophie Morosoli; Laurens Naudts; Natali Helberger; Pablo Cesar

doi:10.1145/3613905.3650750

Transparent AI Disclosure Obligations: Who, What, When, Where, Why, How

Abdallah El Ali, Karthikeya Puttur Venkatraj, Sophie Morosoli, Laurens Naudts, Natali Helberger, Pablo Cesar

Multimedia Computing

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

22 Downloads (Pure)

Abstract

Advances in Generative Artificial Intelligence (AI) are resulting in AI-generated media output that is (nearly) indistinguishable from human-created content. This can drastically impact users and the media sector, especially given global risks of misinformation. While the currently discussed European AI Act aims at addressing these risks through Article 52's AI transparency obligations, its interpretation and implications remain unclear. In this early work, we adopt a participatory AI approach to derive key questions based on Article 52's disclosure obligations. We ran two workshops with researchers, designers, and engineers across disciplines (N=16), where participants deconstructed Article 52's relevant clauses using the 5W1H framework. We contribute a set of 149 questions clustered into five themes and 18 sub-themes. We believe these can not only help inform future legal developments and interpretations of Article 52, but also provide a starting point for Human-Computer Interaction research to (re-)examine disclosure transparency from a human-centered AI lens.

Original language	English
Title of host publication	CHI EA '24
Subtitle of host publication	Extended Abstracts of the 2024 CHI Conference on Human Factors in Computing Systems
Editors	Florian Floyd Mueller, Penny Kyburz, Julie R. Williamson, Corina Sas
Place of Publication	New York, NY
Publisher	ACM
Number of pages	11
ISBN (Electronic)	979-8-4007-0331-7
DOIs	https://doi.org/10.1145/3613905.3650750
Publication status	Published - 2024
Event	2024 CHI Conference on Human Factors in Computing Sytems, CHI EA 2024 - Hybrid, Honolulu, United States Duration: 11 May 2024 → 16 May 2024

Conference

Conference	2024 CHI Conference on Human Factors in Computing Sytems, CHI EA 2024
Country/Territory	United States
City	Hybrid, Honolulu
Period	11/05/24 → 16/05/24

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

Article 52
disclosures
EU AI Act
generative artificial intelligence
law
obligations
research questions
transparency

Access to Document

10.1145/3613905.3650750

3613905.3650750Final published version, 1.01 MB

Cite this

El Ali, A., Venkatraj, K. P., Morosoli, S., Naudts, L., Helberger, N., & Cesar, P. (2024). Transparent AI Disclosure Obligations: Who, What, When, Where, Why, How. In F. F. Mueller, P. Kyburz, J. R. Williamson, & C. Sas (Eds.), CHI EA '24: Extended Abstracts of the 2024 CHI Conference on Human Factors in Computing Systems Article 342 ACM. https://doi.org/10.1145/3613905.3650750

@inproceedings{8908a6c6c099423e9bc4e7d995f99a12,

title = "Transparent AI Disclosure Obligations: Who, What, When, Where, Why, How",

abstract = "Advances in Generative Artificial Intelligence (AI) are resulting in AI-generated media output that is (nearly) indistinguishable from human-created content. This can drastically impact users and the media sector, especially given global risks of misinformation. While the currently discussed European AI Act aims at addressing these risks through Article 52's AI transparency obligations, its interpretation and implications remain unclear. In this early work, we adopt a participatory AI approach to derive key questions based on Article 52's disclosure obligations. We ran two workshops with researchers, designers, and engineers across disciplines (N=16), where participants deconstructed Article 52's relevant clauses using the 5W1H framework. We contribute a set of 149 questions clustered into five themes and 18 sub-themes. We believe these can not only help inform future legal developments and interpretations of Article 52, but also provide a starting point for Human-Computer Interaction research to (re-)examine disclosure transparency from a human-centered AI lens.",

keywords = "Article 52, disclosures, EU AI Act, generative artificial intelligence, law, obligations, research questions, transparency",

author = "{El Ali}, Abdallah and Venkatraj, {Karthikeya Puttur} and Sophie Morosoli and Laurens Naudts and Natali Helberger and Pablo Cesar",

note = "Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.; 2024 CHI Conference on Human Factors in Computing Sytems, CHI EA 2024 ; Conference date: 11-05-2024 Through 16-05-2024",

year = "2024",

doi = "10.1145/3613905.3650750",

language = "English",

editor = "Mueller, {Florian Floyd} and Penny Kyburz and Williamson, {Julie R.} and Corina Sas",

booktitle = "CHI EA '24",

publisher = "ACM",

address = "United States",

}

El Ali, A, Venkatraj, KP, Morosoli, S, Naudts, L, Helberger, N & Cesar, P 2024, Transparent AI Disclosure Obligations: Who, What, When, Where, Why, How. in FF Mueller, P Kyburz, JR Williamson & C Sas (eds), CHI EA '24: Extended Abstracts of the 2024 CHI Conference on Human Factors in Computing Systems., 342, ACM, New York, NY, 2024 CHI Conference on Human Factors in Computing Sytems, CHI EA 2024, Hybrid, Honolulu, United States, 11/05/24. https://doi.org/10.1145/3613905.3650750

Transparent AI Disclosure Obligations: Who, What, When, Where, Why, How. / El Ali, Abdallah; Venkatraj, Karthikeya Puttur; Morosoli, Sophie et al.
CHI EA '24: Extended Abstracts of the 2024 CHI Conference on Human Factors in Computing Systems. ed. / Florian Floyd Mueller; Penny Kyburz; Julie R. Williamson; Corina Sas. New York, NY: ACM, 2024. 342.

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Transparent AI Disclosure Obligations

T2 - 2024 CHI Conference on Human Factors in Computing Sytems, CHI EA 2024

AU - El Ali, Abdallah

AU - Venkatraj, Karthikeya Puttur

AU - Morosoli, Sophie

AU - Naudts, Laurens

AU - Helberger, Natali

AU - Cesar, Pablo

N1 - Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2024

Y1 - 2024

N2 - Advances in Generative Artificial Intelligence (AI) are resulting in AI-generated media output that is (nearly) indistinguishable from human-created content. This can drastically impact users and the media sector, especially given global risks of misinformation. While the currently discussed European AI Act aims at addressing these risks through Article 52's AI transparency obligations, its interpretation and implications remain unclear. In this early work, we adopt a participatory AI approach to derive key questions based on Article 52's disclosure obligations. We ran two workshops with researchers, designers, and engineers across disciplines (N=16), where participants deconstructed Article 52's relevant clauses using the 5W1H framework. We contribute a set of 149 questions clustered into five themes and 18 sub-themes. We believe these can not only help inform future legal developments and interpretations of Article 52, but also provide a starting point for Human-Computer Interaction research to (re-)examine disclosure transparency from a human-centered AI lens.

AB - Advances in Generative Artificial Intelligence (AI) are resulting in AI-generated media output that is (nearly) indistinguishable from human-created content. This can drastically impact users and the media sector, especially given global risks of misinformation. While the currently discussed European AI Act aims at addressing these risks through Article 52's AI transparency obligations, its interpretation and implications remain unclear. In this early work, we adopt a participatory AI approach to derive key questions based on Article 52's disclosure obligations. We ran two workshops with researchers, designers, and engineers across disciplines (N=16), where participants deconstructed Article 52's relevant clauses using the 5W1H framework. We contribute a set of 149 questions clustered into five themes and 18 sub-themes. We believe these can not only help inform future legal developments and interpretations of Article 52, but also provide a starting point for Human-Computer Interaction research to (re-)examine disclosure transparency from a human-centered AI lens.

KW - Article 52

KW - disclosures

KW - EU AI Act

KW - generative artificial intelligence

KW - law

KW - obligations

KW - research questions

KW - transparency

UR - http://www.scopus.com/inward/record.url?scp=85194181995&partnerID=8YFLogxK

U2 - 10.1145/3613905.3650750

DO - 10.1145/3613905.3650750

M3 - Conference contribution

AN - SCOPUS:85194181995

BT - CHI EA '24

A2 - Mueller, Florian Floyd

A2 - Kyburz, Penny

A2 - Williamson, Julie R.

A2 - Sas, Corina

PB - ACM

CY - New York, NY

Y2 - 11 May 2024 through 16 May 2024

ER -

Transparent AI Disclosure Obligations: Who, What, When, Where, Why, How

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this