« Forschung

What does the public think about artificial intelligence?

How does the media report on it?

Public discourse comprises many different ideas and opinions about artificial intelligence. However, decision-makers in politics, business and society should be aware of what people think with regard to shaping technology in such a way that it is compatible with society and democracy. The Opinion Monitor Artificial Intelligence will provide a long-term and reliable picture of the public debate.

In discussions about digitization and artificial intelligence (AI), individual observations of public opinion and opinion published in the mass media are often used to strengthen one’s own position. Though, it is often overlooked that these references are usually not based on systematic and continuous observation of the discourse. Moreover, a look at various demoscopic studies shows that insights into attitudes towards AI are by no means uniform. For example, the Eurobarometer (2017) points to a rather positive attitude of the population towards AI technologies, and a bitkom survey (2018) shows that the majority of respondents believe that AI offers opportunities in innovation. Elsewhere, however, risk perceptions predominate, both in general and in relation to specific fields of application (YouGov 2018).

News

  Research partner: Heinrich Heine University Düsseldorf

  Since 01.04.2021 the project is funded by Stiftung Mercator.

  Since 01.01.2020 – 31.3.2021 the project was funded by the Ministry of Culture and Science of the State of NRW.

News

Here you can find an overview of all events, videos, podcasts and participations of the MeMo:KI team:

.
With the AI Indicator, the Denkfabrik Digitale Arbeitsgemeinschaft shows how to shape policy on the basis of evidence. We are pleased that our project can provide robust data to better understand the effects of AI on the world of work and society.
.

Politicization of AI?

However, the realization of a comprehensive social apprach to AI requires a politicization of the issue. This means first of all that the topic of AI and digitalization be recognized as an object of political debate and decision-making.

Combination of methods

The study combines different empirical methods to observe public and published opinion on AI.

MeMo:KI goals

Besides studying how opinion shifts and how AI is politicized, MeMo:KI pursues three other goals:

Research results

Population survey

You can use the filters to view the results for different age groups as well as genders. To do this, select the appropriate group in the drop-down menu. You can also select multiple filters at the same time (e.g. gender + age group or multiple age groups). The results will update automatically. If you want to reset a filter, simply click again on the previously selected group(s).

Pfeil_rechts

Media coverage

Sample:

Veröffentlichte Artikel aus den marktanteilsstärksten deutschen Print- und Onlinemedien*

Folgende 14 Printmedien werden in der Analyse berücksichtigt: BILD, Süddeutsche Zeitung, Stuttgarter Zeitung Anzeigengemeinschaft, Hannoversche Allgemeine Zeitung, Frankfurter Allgemeine Zeitung, DIE WELT, Kölner Stadt-Anzeiger/Kölnische Rundschau, VRM Tageszeitungen, Rheinische Post, Rhein-Main-Zeitung, Nürnberger Nachrichten, Südwest Presse, Die Rheinpfalz, taz.

Folgende 18 Onlinemedien werden in der Analyse berücksichtigt: chip.de, focus.de, spiegel.de, bild.de, computerbild.de, welt.de, heise.de, giga.de, stern.de, zeit.de, sueddeutsche.de, heute.de, faz.net, rtl.de, pcwelt.de, netzwelt.de, tagesschau.de, pro7.de

Für folgende Printmedien liegt (noch) kein Archivzugang vor und konnten deshalb nicht berücksichtigt werden: Funke Medien (WAZ), Münchner Merkur, Neue Osnabrücker Zeitung, Mediengruppe Thüringen, Zeitungsgruppe Rhein-Neckar / Mannheimer Morgen, Rhein-Zeitung, Schleswig-Holstein-Presse.

*Grundlage bilden die 20 bzw. 18 markanteilsstärksten Print- und Onlinenachrichtenmedien in Deutschland (Quelle: Medienvielfaltsmonitor 2019-I).

Categories – Topics:

General classification of AI: General reports (“The AI”) or overview articles on various aspects of AI.

Methods and technical functionalities of AI: Technical description of AI functions and methods of use e.g. machine learning or recommendation systems.

AI and ethics: reports focusing on ethical aspects, e.g. discrimination, controllability or fairness.

AI in consumer electronics and smart home: use of AI in consumer electronics or in everyday life, e.g. apps, AI in TV or AI in computer games.

Use of AI in business processes: Use of AI in specific business areas, e.g., production processes or personnel recruitment.

AI economy: reports on the tech industry or economic competitiveness, such as start-ups or the presentation of company figures.

AI as a tool and subject of scientific research: reports on AI in research as well as AI as a result or method of scientific research.

Use of AI in medical practice and care: the use of AI in diagnostics and care (e.g. nursing robots) as well as in medical practice (e.g. for skin cancer detection).

AI in education and AI skills: Reports on AI in the education sector (in schools or universities) or around training courses on how to use AI.

AI in the context of ecology, environment and climate: application of AI in relation to the environment, e.g. for disaster warning, as well as the impact of AI on the environment, e.g. through higher energy costs.

Use of AI in police, security agencies, and the judiciary: use of AI in law enforcement (e.g. to monitor public places) or the use of AI in courts (e.g. to search files).

AI law: the concrete enforcement of law as well as the development of legal regulations or legislative proposals.

Policy measures for AI design: reports on government subsidies, political party positions on AI, or concrete government AI strategies.

Use of AI in politics and public administration: integration of AI in administrative processes (e.g. as chatbot) or in election campaigns.

AI as the subject of film, art and literature: articles in which AI is the subject of books, films and games or AI is the subject of (art) exhibitions.

Use of AI in the production of cultural goods: Reports of AI generating culture itself, e.g., painting pictures or composing plays.

AI in journalism: use of AI as a journalistic tool or for research purposes.

AI in interpersonal relationships and as a substitute for interpersonal relationships: Use of AI in direct contact with humans e.g. robots as companions for lonely people or to support interpersonal processes, as in dating apps.

AI and military: articles on AI use in the military or military competitiveness, e.g. autonomous weapons systems, drones and defense systems.

AI in transport and mobility: reports on the use of AI in the transport system. This includes autonomous driving, assistance systems, traffic control planning, or use in passenger transportation (e.g., rail, bus).

Other topics: No clear assignment of the main topic possible.

Twitter Analyse

Literature

bitkom (2018). Künstliche Intelligenz: Bundesbürger sehen vor allem Chancen. Verfügbar unter https://www.bitkom.org/Presse/Presseinformation/Kuenstliche-Intelligenz-Bundesbuerger-sehen-vor-allem-Chancen

de Wilde, P. (2011). No Polity for Old Politics? A Framework for Analyzing the Politicization of European Integration. Journal of European Integration, 33 (5), 559-575.

European Commission and European Parliament (2017). Eurobarometer 87.1. GESIS Data Archive, Cologne. ZA6861 Data file Version 1.2.0, https://doi.org/10.4232/1.12922

Schattschneider, E.E. (1957). Intensity, Visibility, Direction and Scope. American Political Science Review, 51 (4), 933-942.

YouGov (2018). Künstliche Intelligenz: Deutsche sehen eher die Risiken als den Nutzen. Verfügbar unter https://yougov.de/news/2018/09/11/kunstliche-intelligenz-deutsche-sehen-eher-die-ris/

Publications

Kieslich, K., Keller, B., & Starke, C. (2021). AI-Ethics by Design. Evaluating Public Perception on the Importance of Ethical Design Principles of AI. Pre-Print verfügbar auf arXiv: https://arxiv.org/abs/2106.00326

Invited lectures

Marcinkowski, F. (2021, Juni). Monitoring Public Opinion on AI. Vortrag im Seminar AI and the Common Sense, Metz (17.-18.06.2021).

Kieslich, K. & Došenović, P. (2021, Januar). Human-centered or economically driven? Monitoring public opinion on AI in Germany. Gastvortrag im London PUS Seminar, London (27.01.2021).

Presentations at professional conferences

Kieslich, K., Došenović, P., Marcinkowski, F., & Keller, B. (2021, September). Artificial Intelligence: A promising future or a serious threat? An investigation into media coverage and public perceptions of AI in Germany. Präsentation in der Audience and Receptions Studies Section auf ECREA´S 8. European Communication Konferenz in Braga (07.-09.09. 2021).

MeMo:KI-Team

Project management

Research

Science Communication

Researchers involved

Student assistants

Project management

Prof. Dr. Frank Marcinkowski

Prof. Dr. Frank Marcinkowski

Research

Pero Dosenovic

Pero Došenović

Kimon Kieslich

Kimon Kieslich

Researchers involved

Birte Keller

Birte Keller

Dr. des. Christopher Starke

Dr. Christopher Starke

Science Communication

Esther Laukötter

Esther Laukötter

Student assistants

Ole Minuth

Yannick Zelle