Content uploaded by Alexander Pfeiffer
Author content
All content in this area was uploaded by Alexander Pfeiffer on Oct 07, 2020
Content may be subject to copyright.
Proceedings of the
European Conference on the Impact of
Artificial Intelligence and Robotics
ECIAIR 2020
Supported By
Instituto Universitário de Lisboa (ISCTE-IUL)
Portugal
Edited by
Florinda Matos
22 – 23rd October 2020
Copyright The Authors, 2020. All Rights Reserved.
No reproduction, copy or transmission may be made without written permission from the individual authors.
Review Process
Papers submitted to this conference have been double-blind peer reviewed before final acceptance to the
conference. Initially, abstracts were reviewed for relevance and accessibility and successful authors were invited
to submit full papers. Many thanks to the reviewers who helped ensure the quality of all the submissions.
Ethics and Publication Malpractice Policy
ACIL adheres to a strict ethics and publication malpractice policy for all publications – details of which can be
found here:
http://www.academic-conferences.org/policies/ethics-policy-for-publishing-in-the-conference-proceedings-of-
academic-conferences-and-publishing-international-limited/
Conference Proceedings
The Conference Proceedings is a book published with an ISBN and ISSN. The proceedings have been submitted to
a number of accreditation, citation and indexing bodies including Thomson ISI Web of Science and Elsevier
Scopus.
Author affiliation details in these proceedings have been reproduced as supplied by the authors themselves.
The Electronic version of the Conference Proceedings is available to download from
DROPBOX https://tinyurl.com/ECIAIR20 Select Download and then Direct Download to access the Pdf file.
Free download is available for conference participants for a period of 2 weeks after the conference.
The Conference Proceedings for this year and previous years can be purchased from http://academic-
bookshop.com
E-Book ISBN: 978-1-912764-73-0
Book version ISBN: 978-1-912764-74-7
Published by Academic Conferences International Limited
Reading
UK
www.academic-conferences.org
i
Contents
Paper Title Author(s) Page
No
Preface iii
Committee iv
Biographies vii
Research papers
Artificial Intelligence Supported Cognitive
Behavioral Therapy for Treatment of Speech
Anxiety in Virtual Reality Environments
Fredrik Åhs, Peter Mozelius and
Felix Dobslaw
1
Cultural Challenges of the Malicious Use of
Artificial Intelligence in Latin American Regional
Balance
Raynel Batista, Oscar Villar,
Héctor González and Vladimir
Milián
7
The Threat of Using Advanced Technologies by
Terrorists: Psychological Aspect
Darya Bazarkina 14
Eliciting Personal Attitude Changes on Predictive
Policing Based on a Multilinear Narrative
Manuel Brunner, Elke Brucker-
Kley and Thomas Keller
21
Creation of Artificial Bureaucrats Justin B. Bullock and Kyoung-
cheol Kim
30
Mathematical Model of Human Decision: A
Methodological Basis for the Functioning of the
Artificial Intelligence System
Viacheslav Burlov 38
On Handling Ethical Dilemmas in Artificial
Intelligence Systems
Jim Q. Chen 49
The Impact of Digitalization on the Importance
of Intangibles for the Competitive Success in
Medium-Sized Companies
Helmut Döring 58
IEEE P7010-2020 Standard: Use Cases in Ethical
Impact on Human Wellbeing Studies
Alice Vo Edwards 64
User-Centric AI in Public Administration Jan Etscheid 70
Using Text Mining to Analyse Digital
Transformation Impact on People
Florinda Matos, Valter Vairinhos
and Ana Josefa Matos
78
Relevant Distinctions in Relation to
Explainability in the Public Sector
Hanne Marie Motzfeldt and Ayo
Næsborg-Andersen
86
Deepfakes as the New Challenge of National and
International Psychological Security
Konstantin A. Pantserev 93
ii
Paper Title Author(s) Page
No
Malicious Use of Deepfakes and Political
Stability
Evgeny Pashentsev 100
Ethical Challenges in Collecting and Analysing
Biometric Data
Anabela Pereira 108
Introducing the Concept of Digital-Agent
Signatures for Human-Robot-Robot-Human
Interaction
Alexander Pfeiffer, Alesja
Serada, Mark Bugeja, Stephen
Bezzina, Thomas Wernbacher
and Simone Kriglstein
115
Forewarned is Forearmed and Other Risk
Scenarios
Olga Polunina 123
Working Time and Digital Transition: A Complex
and Ambiguous Relationship
Glória Rebelo, Eduardo Simões
and Isabel Salavisa
128
Robotic Process Automation as a Change Agent
for business Processes: Experiences and
Expectations
Juha Saukkonen, Kirsi Kemell,
Maija Haaranen and Erica Svärd
136
Politics of Technology or Technology of Politics? Vincenzo Scalia and Caroline
Stockman
146
“Like 9/11 on Steroids“: AI in the age of
Coronavirus
Keith Scott 153
Robotics and Artificial Intelligence (R&AI)
Perceptions of Consumers and Producers: An
International Comparison Among Portugal and
Spain
Eduardo Tomé, Olga Rivera,
David Lopez and Pedro Soares
de Mello
161
Utilization of Sophisticated Cryptography
Methods in Providing Security in Cyber-physical
Context
Petri Vähäkainu, Martti Lehto
and Antti Kariluoto
169
Blockchain, Facebook and a Polygraph Paulo Vieira, Caroline Stockman
and Paul Crocker
178
Explainability is not Enough: Requirements for
Human-AI-Partnership in Complex Socio-
Technical Systems
Toni Waefler and Ute Schmid 185
Phd Research Paper 195
How Artificial Intelligence Affect the Labour
Market in Poland
Katarzyna Koput 197
Work In Progress Paper 204
Figure Follow: A Step by Step Liberating Device José Cavaleiro Rodrigues, Tiago
Dias and Catarina Silva
206
iii
ECIAIR Preface
These proceedings represent the work of contributors to the 2nd European Conference on the
Impact of Artificial Intelligence and Robotics (ECIAIR 2020), hosted by ACI and Instituto
Universitário de Lisboa (ISCTE-IUL), Portugal on 22-23 October 2020. The Conference Chair is
Dr Florinda Matos, and the Programme Chairs are Dr Ana Maria de Almeida and Prof Isabel
Salavisa, all from Instituto Universitário de Lisboa (ISCTE-IUL), Portugal.
ECIAIR is now a well-established event on the academic research calendar and now, in its 2nd
year, the key aim remains in the opportunity for participants to share ideas and meet people
who hold them. The conference was due to be held at Instituto Universitário de Lisboa (ISCTE-
IUL), Portugal, but because of the global Covid-19 pandemic, it was moved online as a virtual
event. The subjects covered in the papers illustrate the wide range of topics that fall into this
important and ever-growing area of research.
The opening keynote presentation is given by Prof. Mário Figueiredo, from University of
Lisbon, Portugal, on the topic of “Artificial Intelligence: Historical Aspects, Modern
Applications, and Implications”. The second day of the conference will be open by Prof. Jean-
Gabriel Ganascia, Université Pierre et Marie Curie (UPMC), France and a member of the
Institut Universitaire de France, France, who will talk about “Why do we need Ethics and not
just Regulations in AI and Robotics?”.
With an initial submission of 60 abstracts, after the double blind, peer review process there
are 25 academic research papers, 1 PhD research paper, and 1 work-in-progress paper
published in these Conference Proceedings. These papers represent research from Brazil,
Cuba, Denmark, Finland, Germany, Poland, Portugal, Russia, Sweden, Switzerland, UK, USA.
We hope you enjoy the conference.
Dr Florinda Matos
Instituto Universitário de Lisboa (ISCTE-IUL)
Portugal
October 2020
iv
ECIAIR Conference Committee
Dr Kareem Kamal A.Ghany, Beni-Suef University , Egypt; Prof Azween Abdullah, Taylors University,
Malaysia; Dr Kinaz Al Aytouni, Arab International University, Syria; Dr Hanadi AL-Mubaraki, Kuwait
University, Kuwait; Prof Hamid Alasadi, Iraq University college, Iraq; Prof Laurice Alexandre, Sorbonne
Paris Cité , France; Dr José Álvarez-García, University of Extremadura, Spain; Dr Xiomi An, Renmin
University of China, China; Prof Antonios Andreatos, Hellenic Air Force Academy, Greece; Prof Oscar
Arias Londono, Institucion Universitaria de Envigado, Colombia; Dr Gil Ariely, Interdisciplinary Center,
Herzliya, Israel; Dr Sotiris Avgousti, Cyprus University Of Technology, Cyprus; Prof Rosalina Babo,
Polytechnic of Porto, Porto Accounting and Business School, Portugal; Prof Liz Bacon, Abertay
University, UK; MSc Jordi Bieger, TU Delft, Netherlands; Prof Elias Carayannis, George Washington
University, USA; Prof Davide Carneiro, Polytechnic Institute of Porto, Portugal; Prof Karen Cham,
University of Brighton, UK; Prof Jim Chen, U.S. National Defense University, U.S.A.; Dr Pericles Cheng,
European University Cyprus, Cyprus; Prof Koteshwar Chirumalla, Malardalen University, Sweden; Mr.
David Comiskey, Ulster University, UK; Dr Leonardo Costa, Universidade Católica Portuguesa - Católica
Porto Business School, Portugal; Prof Carmen-Eugenia Costea, The Bucharest University of Economic
Studies, Romania; Prof Larry Crockett, Augsburg University, USA; Dr Marija Cubric, University of
Hertfordshire, UK; Francesca Dal Mas, Università di Udine/Università la Sapienza, Italy; Assc Ben
Daniel, University of Otago, New Zealand; Prof Justine Daramola, Cape Peninsula University of
Technology, South Africa; Geoffrey Darnton, WMG, University of Warwick, UK; Mr Martin De Bonis,
Alma Mater Studiorum, Italy; Prof Armando Carlos de Pina Filho, Federal University of Rio de Janeiro
(UFRJ), Brazil; Dr Martin De Saulles, University of Brighton, UK; Dr María de la Cruz Del Río-Rama,
University of Vigo, Spain; Dr Souâd Demigha, CRI Univ of Paris 1 La Sorbonne, France; Paolo Di Muro,
Politecnico di Milano School of Management, Italy; Dr Mihaela Diaconu, ”Gheorghe Asachi” University
of Iasi, Romania; Inês Domingues, DEIS-ISEC, Portugal; Dr Patricio Domingues, Polytechnic Institute of
Leiria, Portugal; Prof Yanqing Duan, University of Bedfordshire, UK; Prof. John Edwards, Aston Business
School, UK; Dr Kelechi Ekuma, The Global Development Institute, University of Manchester, UK; Dr
Scott Erickson, Ithaca College- School of Business, USA; Dr José Esteves, IE Business School, Spain; PhD
Fernanda Faini, CIRSFID - University of Bologna, Italy; Dr Georgios Fessakis, University of the Aegean,
Greece; Prof Eric Filiol, ENSIBS, Vannes, France & CNAM, Paris, France; Dr Panagiotis Fotaris, University
of Brighton, UK; Prof Andreas Giannakoulopoulos, Ionian University, Greece; Dr Valerie Priscilla Goby,
Zayed University, United Arab Emirates; Dr Amol Gore, UAE Government HCT, UAE; Dr Paul Griffiths,
Ecole de Management de Normandie, Oxford, UK; Dr Hossein hakimpour, IAU , Iran; Prof William Halal,
George Washington University, USA; Prof Ali Hessami, City University London, UK; Dr Grant Howard,
University of South Africa (Unisa), South Africa; Prof Ulrike Hugl, Innsbruck University, Faculty of
Business and Management, Department of Accounting, Auditing and Taxation, Austria; Prof Hamid
Jahankhani, Northumbria University London, UK; Dr Aman Jatain, Amity University, India; Dr Runa
Jesmin, University of Roehampton, UK; Dr Jari Jussila, Häme University of Applied Sciences, Finland; Dr
Selvi Kannan, Victoria University, Australia; Prof Ergina Kavallieratou, University of the Aegean,
Greece; Dr Harri Ketamo, Headai ltd, Finland; Dr Nasrullah Khilji, University of Bedfordshire, UK; Prof
Tatiana Khvatova, Peter the Great St. Petersburg Polytechnic Universtity, Russia; Prof Jesuk Ko,
Universidad Mayor de San Andres, Bolivia; Prof Michael Kohlegger, Institute for Web Technologies &
Applications, Austria; Prof Renata Korsakiene, Vilnius Gediminas Technical University, Lithuania; Prof
Ibrahim Krasniqi, University of Peja, Kosovo; Prof Konstadinos Kutsikos, Business School, University of
the Aegean, Greece; Dr Jean Lai, Hong Kong Baptist University, Hong Kong; Dr Isah Abdullahi Lawal,
Noroff University College, Norway; Dr Efstratios Livanis, University of Macedonia, Greece; Prof Eurico
Lopes, Instituto Politécnico de Castelo Branco, Portugal; Dr Martin Magdin, Constantine the
Philosopher University in Nitra, Faculty of Natural Sciences, Department of Informatics , Slovakia; Dr
Paolo Magrassi, Alephuture, Switzerland; Dr Hossein Malekinezhad, Islamic Azad University, Naragh
Branch, Iran; Prof António Martins, Universidade Aberta, Portugal; Prof Maurizio Massaro, Ca' Foscari
University of Venice, Italy; Dr Nuno Melão, Polytechnic Institute of Viseu, Portugal; Prof Anabela
v
Mesquita, Polytechnic of Porto, Portugal; Prof David Methé, Institute of International Strategy at Tokyo
International University in Tokyo, Japan; Dr Larisa Mihoreanu, ANMDM Bucharest, Romania; Dr
Clemente Minonne, Lucerne University of Applied Sciences, Institute for Innovation and Technology
Management, Switzerland; Prof Harekrishna Misra, Institute of Rural Management Anand , India; Assc
Ludmila Mladkova, University of Economics Prague, Faculty of Business Administration, Czech
ZĞƉƵďůŝĐ͖ ƌ ƌƚƵƌ DŽĚůŝŷƐŬŝ͕ hŶŝǀĞƌƐŝƚLJ ŽĨ BſĚǍ - Faculty of Management, Poland; Prof Fernando
Moreira, Universidade Portucalense, Portugal; Prof John Morison, Queen's University Belfast , UK; Dr
Rabeh Morrar, Northumbria University, UK; Dr Hafizi Muhamad Ali, Yanbu University College, Saudi
Arabia; Dr Antonio Muñoz, Universidad de Málaga, Spain; Prof Mihaela Muresan, Dimitrie Cantemir
Christian University, Romania; Dr Minoru Nakayama, Tokyo Institute of Technology, Japan; Dr MItt
Nowshade Kabir, Trouvus, USA; Prof Beatrice Orlando, Sapienza University of Rome, Italy; Prof Evgeny
Pashentsev, Diplomatic Academy at the Ministry of Foreign Affairs of the Russian Federation, Russia;
Assc Corina Pelau, Bucharest University of Economic Studies,, Romania; Dr Parag Pendharkar, Penn
State Harrisburg , USA; Dr Alexander Pfeiffer, Applied Game Studies, Austria; Dr Cosmin Popa,
Innovative Agricultural Services, UK; Prof Ricardo Queirós, ESMAD/P.Porto, Portugal; Prof Carlos
Rabadão, Politechnic of Leiria, Portugal; Assc Liana Razmerita, Copenhagen Business School, Denmark;
Dr Marcin Relich, University of Zielona Gora, Poland; Prof José Carlos Ribeiro, Polytechnic Institute of
Leiria, Portugal; Prof Sandra Ribeiro, ISCAP, IPP-Porto, Portugal; Dr Martin Rich, Cass Business School,
UK; Dr Kenneth Rogerson, Sanford School of Public Policy, USA; Prof Göran Roos, University of South
Australia, Australia; Dr Eleni Rossiou, Experimental School of Aristotle University , Greece; Prof Neil
Rowe, U.S. Naval Postgraduate School, USA; Dr Melissa SAADOUN, EIVP, France; Prof Lili Saghafi,
MSMU, USA; Prof Mustafa Sagsan, Near East University, Turkish Rebuplic of Northern Cyprus; Prof
Abdel-Badeeh Salem, Faculty of Computer and Information Sciences, Ain Shams University, Cairo,
Egypt; Dr Char Sample, US Army Research Laboratory, USA; Dr Navjot Sandhu, Birmingham City
University, UK; Prof Vitor Santos, NOVA IMS - New University of Lisbon, Portugla; Prof Ramanamurthy
Saripalli, Pragati Engineering College, India; Prof Markus Schatten, Faculty of Organization and
Informatics, University of Zagreb, Croatia; Dr Elena Serova, National Research University Higher School
of Economics , Russia; Assistant Professor Sandro Serpa, Universidade dos Açores, Portugal; Dr Armin
Shams, Sharif University of Technology, Iran; Dr Yilun Shang, Northumbria University, UK; Dr Eric Shiu,
University of Birmingham, UK; Prof Fernando Silva, Polytechnic Institute of Leiria, Portugal; Prof
Andrzej Sobczak, Warsaw School of Economics, Poland; Dr Caroline Stockman, University of
Winchester, UK; Dr Darijus Strasunskas, Hemit, Norway; ƌ KůŐĂ ^ƚƌŝŬƵůŝĞŶĦ͕ <ĂƵŶĂƐ hŶŝǀĞƌƐŝƚLJ ŽĨ
Technology, Lithuania; Dr Marta-Christina Suciu, Bucharest University of Economic Studies, România;
Dr Saloomeh Tabari, Sheffield Hallam University, UK; Prof Ramayah Thurasamy, Universiti Sains
Malaysia, Malaysia; Assc Milan Todorovic, Royal Melbourne Institute of Technology, Australia; Prof
Jim Torresen, University of Oslo, Norway; Dr khan ferdous wahid, Airbus, Germany; Prof Fang Wang,
Nankai University, China; Prof Murdoch Watney, University of Johannesburg, South Africa; Prof Bruce
Watson, Stellenbosch University, South Africa; Dr Santoso Wibowo, Central Queensland University,
Australia; Prof Robert J. Wierzbicki, University of Applied Sciences Mittweida, Deutschland; Dr Marcus
Winter, University of Brighton, UK; Dr Adam Wong, School of Professional Education & Executive
Development, The Hong Kong Polytechnic University, Hong Kong; Mr Jason Wong, KDU University
College, Malaysia; Mr Tuan Yu, Kent Business School, UK; Prof Daiva Zostautiene, Kaunas University of
Technology, Lithuania;
Introducing the Concept of Digital-Agent Signatures for Human-
Robot-Robot-Human Interaction
Alexander Pfeiffer1.2,3, Alesja Serada7, Mark Bugeja3, Stephen Bezzina6, Thomas
Wernbacher2 and Simone Kriglstein4 5
1Massachusetts Institute of Technology (MIT), Cambridge, MA, USA
2Center for Applied Game Studies, Donau-Universität Krems (DUK), Austria
3University of Malta (UoM), Msida, Malta
4Austrian Institute of Technology GmbH (AIT), Vienna, Austria
5University of Vienna, Vienna, Austria
6Ministry for Education and Employment, Floriana, Malta
7University of Vaasa, Finland
alex_pf@mit.edu
alesja.serada@uwasa.fi
mark.bugeja@um.edu.mt
mail@stephenbezzina.com
Thomas.wernbacher@donau-uni.ac.at
simone.kriglstein@ait.ac.at
DOI: 10.34190/EAIR.20.004
Abstract: Digital/electronic identities are essential components of collaborative robots/robots and human-robot/robot-
human interactions. Through such identities, digital agents (AI powered software or robots/bots) are entrusted with tasks
in the name of certain individuals/companies. Digital identities can come from various sources; these can be assigned by an
employer, through a service provided by a government entity or an external company specializing in the creation of such
signatures or generated through an interface like Facebook Connect. All these different sources offer a range of varying
levels of trust, both within the institution where the signature is principally used, but especially when interacting with third
parties. Ultimately, this level of trust or its valuation is a determining factor in how far the authorization of the respective
digital/electronic signature goes. The authors describe the application of digital/electronic signatures from human
employees or legal entities which, simultaneously with the main task, generate sub-signatures for the respective digital
agent.The topic is presented from a technical perspective as well as from a social science point of view.
Keywords: Digital Agents, Digital Identity, Self-Sovereign Identity, Blockchain, AI
1. Introduction
A new high-class gaming PC bought by an 11-year-old via Alexa voice command; a reservation in a luxury
restaurant, unintentionally made by google-assistant; or a 1000 Euro tax overcharge, due to an error in the AI-
assisted accounting software; cases, which are solved nowadays due to well-written terms and conditions, a
helpful service hotline or in the worst case in court.
Now imagine these kinds of problems on a larger scale. An unauthorized person orders production machines
via an AI-assisted purchasing software; a digital assistant books an unplanned business trip without knowing
who gave the order initially; a minor software error that miscalculates the tax by a "0", or production machines
that have not been operated in accordance with their intended use.
Most of these cases are unlikely to be solved on a goodwill basis but will have more significant consequences.
Often, however, one will have to ask first who should be held responsible at all. The ever-faster technological
progress is therefore not only a blessing (e.g. through cost savings due to efficient production), but also brings
problems with it. Problems which from a social, ethical, political, technological or legal we have not yet
managed to solve.
Triggered by the current Covid-19 situation, automation and production using machines with AI elements are
being discussed even more intensively and pursued more rapidly. (BBC Article 2020)1
1 https://www.bbc.com/news/technology-52340651 (
115
Alexander Pfeiffer et al
To put the problem into perspective: We currently lack a trustworthy solution to show orders given to "AI-
assisted Software Solutions", "AI-driven language assistants", to AI-assisted robots/bots" between all parties
involved. The problem becomes more prominent, the more partners are involved and whether they go beyond
companies, municipalities, countries or even continents.
Trust is described by Jøsang (2016) as a subjective belief in the reliability, honesty and security of an entity on
which we depend for our welfare, and these entities contain software, hardware, data, people and
organizations. Two components might be needed to create this trust in the digital space: digital/electronic
identities and storage of data and processes on data hubs that are accepted by all parties involved.
2. Research Questions
The authors pursue the following research questions:
x How can secured digital identities be transferred to AI agents?
x What role can Blockchain technologies in connection with different forms of ID-verification play?
x Which aspects must be considered in the ethics debate, especially when AI and blockchain takes over
our activities in a more complex (though still human-defined) framework.
3. Aim and Methodology
We reviewed the literature on existing solutions and discussions on the topic of digital agents, virtual
environments, self-sovereign identity, qualified digital signatures, ethical discussions related to virtual
world/agents and blockchain. In our research, we stress the importance of performative, multiple and
adjustable digital identities that can be constructed (or generated as sets of signatures and sub-signatures) and
controlled in a way similar to avatars in a virtual environment.
4. Related Research
Goodell & Aste (2019) suggest that potential users of digital identification systems should be free to operate
several instances of identities, each suited for a specific aim. The Authors provided a general blueprint for
'trustless' interactions with multiple identifiers but did not extend their concept to specific use cases. In this
paper, this idea is taken one step further by providing an actual technical embodiment of a similar idea.
The design of digital identities in virtual worlds provides another fruitful perspective. Full anonymity and
inconsistency of a player's identity between play sessions, provoked online abuse, even in the earliest virtual
worlds (McDonough 1999). However, as McDonough (1999) shows, making all interactions between players
open and public lead to players' discomfort and protective behaviors to restore at least some level of privacy.
Thus, a stable identity in a digital world is required to create trust, but a person should also be able to project
different facets of this identity to different actors, just as we take up different roles in social interactions.
Qualitative studies of online identities and privacy management in social networks produce the same
conclusion. In general, digital ethnographers have successfully demonstrated that it would be a mistake to
reduce relationships between real-world identities and online personas to direct, one-to-one connection
(Marwick and Boyd 2014, Bancroft and Scott Reid 2017).
5. Blockchain
Blockchain can play a key role in the non-manipulable and trusted storage and application of digital identities,
their transfer to digital agents and the recording of tasks performed by those. Blockchain Systems as we know
them today are based on the white paper “Bitcoin: A Peer-to-Peer Electronic Cash System”, by the anonymous
author Satoshi Nakamoto (2008). Blockchain technologies belong to the Distributed Ledger Systems or DLTs in
short. DLTs work through different computers that store information of the same type. The ledger is therefore
divided into different locations, operated by different persons or companies, none of these people or
companies has to know or personally agree with each other when using a public Blockchain. Blockchains are
unique due to the way they operate, which is based upon a set of rules. These rules vary slightly depending on
the Blockchain system used. Transactions are then combined in a block and stored in encrypted form. This
process is intended to ensure that the same information is actually stored on the distributed systems and that
there is no file or text information among them that may have the same file name and size as all the others but
does not contain the correct information. The storage process of a Blockchain is, therefore based on the fact
that new data blocks are continuously generated. Each of these new entries (blocks) increases the size of the
Blockchain.
116
Alexander Pfeiffer et al
Blockchain systems can operate in three different ways:
x Private Blockchain: is a closed system and is operated exclusively within organizations, companies or
government structures. No information is passed on to the outside world unless there is evidence that
a transaction has taken place.
x Blockchain operated by a consortium: serves connected parties who have a common goal. Consortium
partners may join the Blockchain, based on joint agreements.
x Public Blockchain: has no restrictions on joining or leaving the Blockchain. All information is public,
although it is possible to store some information in encrypted form.
x Private and consortium Blockchains can also store information on a public Blockchain, for example, the
hash value of all transactions within 24 hours. This keeps the data content itself private but ensures
that no data manipulation takes place retroactively. Not block by block, but still, as in the example
above, for all data older than 24 hours
6. Digital Identities
Digital/electronic identities are essential components of collaborative robots/robots and human-robot/robot-
human interactions. Through such identities, digital agents (AI powered software or robots/bots) are entrusted
with tasks in the name of certain individuals/companies. Digital identities can come from various sources;
these can be assigned by an employer, through a service provided by a government entity (For example
signatures that comply with the EIDAS regulation2) an external company specializing in the creation of such
signatures, the self-sovereign identity (SSI) movement (Sovrin Foundation) or generated through an interface
like Facebook Connect. All these different sources offer a range of varying levels of trust, both within the
institution where the signature is principally used, but especially when interacting with third parties.
Ultimately, this level of trust or its valuation is a determining factor in how far the authorization of the
respective digital/electronic signature goes.
The first state-supported pilot project for a digital identity on blockchain in the EU was launched in Zug,
Switzerland, in September 2017 (Blockchain-Identität für alle Einwohner 2017). It is based on the Ethereum
blockchain. In June 2018 these blockchain identities were officially used for voting (Eixelsberger et al. 2019,
514).
Another type of projects can be seen as "data cooperatives" described by Giannopoulou (2020): they approach
"data as a common value" and create tools for its collective regulation. However, community standards for
data management in such projects remain opaque. If closed ecosystems of data emerge as a result, abuse and
exploitation within them are technically viable. A non-authoritarian way to manage digital identities is to
provide as many opportunities for integration as possible.
7. Digital Interaction & the role of digital agents
We can distinguish between three different types of interaction:
1. Human with computer interaction: The average person logs in 7-25 times per day (Greene). In the
simplest form of a login system used. This can either be assigned by a system administrator (human or
software) or by the user himself.
2. Computer to computer interaction: (digital-agent with digital-agent, digital-agent with software,
software with digital-agent). In this case, too, a digital identity in the sense of proof of entitlement
must be provided, in the best-case scenario, this process can track right back to the original source. It
is essential, however, that the instruction to the software or digital-agent is guaranteed by the most
secure authentication possible.
3. Computer to human interaction: This takes place when a digital-agent, approaches a human to enter
further data, to perform a production step or to mark work as completed. Like the previous case, it is
also vital that the system can trace from which source, or from which sources, the initial order
originated.
Wooldridge points out that there is no generally agreed definition of agents. In 2000 he proposed the
following definition, which reflects a revision of his thoughts from 1995. Wooldridge:
2 https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=uriserv%3AOJ.L_.2014.257.01.0073.01.ENG
117
Alexander Pfeiffer et al
“An agent is a computer system that is situated in some environment, and that is capable of autonomous
action in this environment in order to meet its design objectives.”
But he also says that this definition does not yet reflect the degree of autonomy of an agent nor the space in
which it is located.
Monostori et al. describe the role of agents along with other factors as follows. Digital agents should:
x Have a purpose to fulfill,
x Perform autonomous behavior and control both of their actions within the environment,
x Perform real-time information processing and adapt themselves to new situations,
x Prioritize events in accordance with their preferences,
x Exhibit intelligence, to some degree, from applying fixed rules to reasoning, planning, and learning
capabilities,
x Interact with their environment in which they are operating, including the interaction with other
agents,
x Be adaptive, that is, capable of tailoring their behavior to the changes of the environment without the
intervention of their designer,
x Work as genuinely and transparently as possible, and
x Be credible and trustworthy in providing information to others.
Obviously, spaces can exist where more than one agent exists. Huhns and Stephens describe the conditions of
such environments:
x Multiagent environments provide an infrastructure specifying communication and interaction
protocols.
x Multiagent environments are typically open and have no centralized designer.
x Multiagent environments contain agents that are autonomous and distributed, and may be self-
interested or cooperative.
Burden and Savin-Baden (2019) define four different types of “AI-Systems”, which can be well adopted for the
thoughts about digital agents:
x Simple Algorithms – probably 99% of most computer programs, even complex Enterprise Resource
Planning (ERP) and Customer Relationship Management (CRM) systems since they are highly linear and
predictable.
x Complex Algorithms – programs such as, but not limited to, machine learning, deep learning, neural
networks, Bayesian networks and fuzzy logic where the complexity of the inner code starts to move
beyond simple linear relationships. Many systems currently referred to as AI sit here.
x Artificial General Intelligence – closer to what the public image of AI is, a system that can be applied to
a wide range of problems and solve them to a better or similar level as a human.
x Artificial Sentience – beloved of science-fiction, code which ‘thinks’ and is ‘self-aware’.
8. Proposal of the E-ID Wallet concept and Digital Agent Signature
How can the various concepts of digital identities described so far be applied to digital agents? And what key
role can Blockchain technologies play? To answer the first research question, the authors would now like to
discuss the concepts of the E-ID wallet and the digital agent signature.
An E-ID wallet - as perceived by the authors - is a wallet for blockchain-tokens, which is linked to one or more
digital signatures of the owner(s). Valid signatures include government-issued signatures, any signature from a
self-sovereign identity app, a signature issued by an educational institution, signatures issued by an identity
verification company or a connection to a social media account. There are different levels of trust in the digital
signature to be considered.
Depending on the selected signature type, the proof of the signature transaction is stored and displayed
differently. The signature hash value can be published, for example, on a protocol page of the respective trust
center, on the blockchain used by the SSI app, in which case, a token (including the private data as encrypted
message) is sent from the (signed E-ID wallet) of the SSI app provider to the newly signed E-ID wallet of the
user.
118
Alexander Pfeiffer et al
The signature chain can be retrieved for each E-ID wallet. For example, if an E-ID wallet is dedicated to a
department of a company, the user will see that the primary account of the company and the person
responsible for the department have signed as well as the user who originally signed the main account of the
company and how the primary person responsible for the department got this status. Whether the private
data is publicly accessible or encrypted is, of course, always subject to the person or institution and their
needs. In other words, whether it is essential that everyone can see whom the wallet is assigned to, or if only
persons, company-partners, or other departments of a company who gain access to this information should
know the ownership.
However, the distinctive feature of the E-ID Wallet is that, in addition to digital identities, it can and should
also hold blockchain-based tokens and can, therefore, be used for utility tokens linked to an identity on the
one hand, but also for cryptocurrencies as a form of payment with proof of identity (to counteract money
laundering and other similar problems) on the other.
Now it is a matter of connecting digital signatures and E-ID wallets with digital agents so that their distribution
of tasks and progress is stored with the highest possible security and allows the digital agents to interact with
third parties.
It should also be noted that some blockchain-token wallets already offer the possibility to name the wallet
publicly (for example original Ardor wallet, but here it is a pure self-authentication). One way of proving the
identity of a blockchain wallet would be to have a digitally signed PDF that specifies the blockchain address
and is digitally signed by the user, with a transaction taking place from the blockchain address that references
the PDF and its hash value together with possible location.
The basic idea of the digital agent signature is to connect the (digital) identity of the user of digital-agent with
the digital agent itself. The user uses a signature that is available in his SSI app. This allows the user to select
the appropriate signature for the different applications. When the digital-agent is instructed to carry out an
administrative task, such as monitoring and paying tax returns, a government-issued signature will be used. If
the digital-agent is instructed to search for and purchase the best possible car insurance, a signature issued by
an identity verification company is used, a digital-agent is instructed to compile and enroll in the best schedule
for study. That signature issued by the university is then used, and if the digital-agent is acting on behalf of the
user on a love mediation platform, an even lower level of verification may be sufficient.
If the signature is used in a work context, a custom SSI-app provided by the company with authorized
signatures, or an SSI-app that the user typically uses is used and where there are one or more signatures to
choose from, is able to verify the identity of the user, the user’s position in the company, the user’s rights
within the company and also the authenticity of the company for which the user is operating can be utilized.
The particular feature of the digital-agent signature is that a token transfer is triggered during the signature
process. From the E-ID wallet of the person or company for whom this person works to the digital-agent's E-ID
wallet. The data can be controlled via a system of shared keys in various levels, which are again connected to
an identity management system.
9. Ethical debate
The discussion on identities for digital agents and their authorizations must not only be conducted from a
technical perspective, but also from an ethical perspective. The authors would now like to address the
different perspectives of the discussion and thus answer the second research question. Identity management
in digital spaces acquired new meanings after the introduction of blockchain technologies. The possibility to
create an indestructible and automatically verifiable personal record gave a new meaning to individual
autonomy online, but it also raised several ethical concerns. Among others, binding a personal identity to a
single non-destructible digital record violates the 'right to be forgotten', which is also a part of the European
General Data Protection Regulation. It contradicts the principle of 'purpose limitation', which states that
personal data should be kept as long as it is required by the purpose of collecting it, but no longer (GDPR).
Privacy, in general, is the recurrent topic in ethical debates on data subjects and digital identities. Earlier
implementations of blockchain such as Bitcoin and later many other cryptocurrencies sought to resolve the
issue of privacy by anonymity. However, such cryptocurrencies as Bitcoin offer pseudonymity, at best, and
119
Alexander Pfeiffer et al
their lack of identity protection has been uncovered in a rather short time (Reid and Harrigan 2012). This
concern leads to the development of more privacy-oriented cryptocurrencies such as Monero and Zcash as
secure alternatives to Bitcoin and Ether. However, despite sophisticated cryptography, some rather intuitive
methods to identify owners of financial assets have been used by Wall Street traders "for decades if not
centuries" (Yermack 2017, 18), and they remain relevant in case of blockchain, as the discovery of the Bitcoin
wallet that allegedly belongs to Satoshi Nakamoto suggests (Voell 2020). Blockchain Developer Wikis, started
to publish tutorials on how to secure sensitive data using a hybrid solution that stores the sensitive data in a
centralized database and places a unique proof of all operations on the public blockchain. (see, e.g.
Ardordocs3)
The promise of anonymity backfired by the association of cryptocurrencies with criminal activities such as drug
trade and money laundering (see Latimer and Duffy 2019 for current evaluation of financial risks related to
cryptocurrencies). However, even communities that put anonymity first tend to operate under somehow
authentic social personas, as the study of the actual 'darknet' bitcoin users by Bancroft and Scott Reid has
shown. Sellers of illicit goods maintain stable pseudonymous identities that function in the same way as
brands, in order to manage their reputation among buyers (Bancroft and Scott Reid 2017). Once again, privacy
is challenged, re-constructed and re-negotiated, and new digital aspects of old identities are generated,
confirmed and managed to enable social and financial interactions online.
The next concern, related to our proposal, is the question of consent. Human consent in information systems
becomes a means "to mediate the expression of autonomy through technological applications" (Giannopoulou
2020) by well informed and self-determining subjects. However, retaining agency through consent in a
technological society becomes dubious. How informed is informed consent in global information systems that
are too complex to understand? Truly informed consent presumes the amount of responsibility and a cognitive
load probably unbearable for a human being who is not a security engineer by occupation. As Langdon Winner
summarizes in his writings on autonomous technology: "With the overload of information so monumental,
possibilities once crucial to citizenship are neutralized" (Winner 2001, 296).
Furthermore, all data subjects involved in electronic communications leave 'digital traces' that can be scrapped
without consent. In the best-case scenario, this data can be used to enable more comfortable coexistence of
humans and non-humans in responsive smart environments, which should be the goal of digital identity
projects. In the worst case, prevented from "forming or formulating a desire" (Rouvroy et al. 2013), a human
agent is deprived of choice, purpose and opportunities for self-actualization, much like in a science fiction film
The Matrix (1999). In reality, scrapping seemingly non-private data to use it for algorithmic decisions on
personal safety have created controversies around AirBnB, among others, for banning marginalized but
otherwise law-abiding users from the service (Dickson 2020; see also Jhaver et al. 2018 on coping behaviors of
AirBnB hosts when 'negotiating' with opaque algorithms).
Another concern, especially in communication between human and non-human agents, arises when a digital
identity is prioritized over the real human being when making an important decision. Consequences can be
grave in case of an algorithmic decision about human matters. In critical information studies, a concept of a
"digital" or a "statistical double" has been introduced, and the potentially repressive rule of algorithms has
been described as "algorithmic dominance" (Giannopoulou 2020) or "algorithmic governmentality" (Rouvroy
et al. 2013).
Artefacts always have ethical values encoded in them (Winner 1980). 'Platform ethics' of blockchains, enforced
by 'smart contracts', can potentially magnify existing biases and power disbalances in electronic systems. Use
of blockchain in the capacity of "a trust machine" does not guarantee fairness. In his discussion of potential
applications of blockchains in corporate governance, David Yermack notes that "the regulations embedded in a
blockchain's software code could favor some participating companies at the expense of others" (Yermack
2015, 27) and stresses the importance of possible human intervention. As an example of such intervention, he
reminds of the DAO hack of the Ethereum platform in 2016: after the hack, 85% of Ethereum miners agreed to
'hard fork' the compromised platform and negated the consequences of illicit behavior.
3 Ardordocs: https://ardordocs.jelurida.com/Securing_sensitive_data_with_the_blockchain
120
Alexander Pfeiffer et al
This example also shows that human judgement should always be prioritized, even – and especially – if not all
human actors agree about the same priorities. The fully automated algorithmic consensus is the scenario that
theoretically leads to 'domination' of robots over humans. If digital identities are prioritized over natural
persons in the system, the consensus in it will likely be in favor of artificial intelligence, and it will probably
exclude specifically human interests from consideration. Biometrics combined with blockchain and AI is yet
another development of this same potentially harmful scenario: while blockchain is immutable and tamper-
proof, a human body is not. A reasonable level of doubt should be guaranteed in every system that combines
indestructible records with potentially flaccid biometric data.
In general, how concerned should we be with artificial agents? Ethical concerns about machine intelligence are
often magnified with the existential fear of achieving 'singularity', the future event that will herald the total
superiority of machine superintelligence over human capacities (Bostrom 2002). However, another vision
appears to be more realistic - a cooperative vision of human and non-human entities who cooperate to reach
common goals set by human actors and system designers (machines ultimately lack goal-setting abilities). In
his review of original ideas of artificial intelligence by Alan Turing and J.C.R. Licklider, Oscar Schwartz shows
how Turing's 'automotive vision' feeds the anxiety of "computers automating and replacing humans". In
contrast, Licklider's "hybrid vision of AI" relies on human-machine collaboration that harnesses the power of
machine intelligence (Schwartz 2018).
10. Conclusion
From a technical perspective: The digital agent signature combines self-sovereign identity (e.g.: digital qualified
signatures) with verified blockchain wallets (E-ID wallets) and non-tradable utility tokens as a carrier medium
for data and authorization to operate. This not only provides a complete record of interactions with and
between digital-agents and their tasks, but it also ensures that you can see who has given the orders for the
actions. All this information can be kept private, either in whole or in part, with the ability to assign shared-
keys for access rights. Only the hash value that a transaction has taken place should be public or at least
shared between the consortium.
From the ethnic perspective: Virtual worlds have taught us that ‘social types’ in the material worlds can be
compared to “real-life avatars” that interact with other human and non-human actors. A game educator James
Paul Gee has argued that a gamer’s self is a unified “sum and intersection” of online and offline identities and
experiences (Gee 2015, 100). This understanding invites us to consider digital identities that are unified and
plural at the same time, as a better fit for realistic, social and respectful implementations of identity
management technologies. A cybernetic model of privacy in electronic networks should be re-evaluated to
correspond to the social, intrinsically contextual way of practicing privacy in interaction with human and non-
human agents. While technical limitations of ‘restricted access constitute the cybernetic model’, the social
model is about ‘shared access’, a dynamic way to establish and negotiate boundaries and connections in
virtual environments.
The authors suggest that understanding identity management for digital agents is similar to how we perceive
avatars. As identities which are stable and consistent; they are attached to a single human person or a legal
entity that can be identified on request. At the same time, this identity solution allows human agents to
control which accounts and personal records to provide in interaction with non-human entities to achieve their
goals in a private and secure, and yet transparent manner.
11. Further Research
The authors propose to consider both aspects together in future research projects. Technical progress should
always be accompanied by a socio-political perspective. The authors would like to pursue this goal in further
research projects. The proposals from this "Vision-Paper" should be put into practice and a prototype of an E-
ID wallet should be created and tested and discussed in different use cases. Different interaction variations
(man to man, man to machine, machine to machine) will be considered and the focus will be to discover
possible fields of problems.
References
Cryptomarket Users. Information, Communication & Society 20, no. 4 (April 3): 497–512.
https://doi.org/10.1080/1369118X.2016.1187643.
121
Alexander Pfeiffer et al
Gee, James Paul. (2014). Unified Discourse Analysis: Language, Reality, Virtual Worlds and Video Games. 1 edition.
London ; New York: Routledge,
Bancroft, A. and Reid P. S. (2017). Challenging the Techno-Politics of Anonymity: The Case of
Blockchain-Identität für alle Einwohner (2017). Stadt Zug. https://www.stadtzug.ch/newsarchiv/383355
Bostrom, N. (2002). Existential Risks. Analyzing Human Extinction Scenarios and Related Hazards. Journal of Evolution and
Technology, Vol. 9, No. 1.
Burden, D.; Savin-Baden, M. (2019) Virtual Humans, Today and Tomorrow, part of: Chapman & Hall/CRC, Artificial
Intelligence and Robotics Series, CRC Press, Taylor & Francis Group, Boca Raton, USA
Dickson, EJ. (2020). Who’s Allowed to Use Airbnb? Rolling Stone (January 8) https://www.rollingstone.com/culture/culture-
news/airbnb-sex-worker-discrimination-935048/
Eixelsberger, W., Wundara, M. und Huemer, W. (2019). Blockchain in der Verwaltung. Handbuch E-Government. Springer
Fachmedien Wiesbaden, pp. 506-518. https://doi.org/10.1007/978-3-658-21402-9_43
GDPR Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of
natural persons with regard to the processing of personal data and on the free movement of such data, and repealing
Directive 95/46/EC (General Data Protection Regulation) http://data.europa.eu/eli/reg/2016/679/2016-05-04
Giannopoulou, A. (2020). Algorithmic Systems: The Consent Is in the Detail? Internet Policy Review 9, no. 1 (March 23).
https://policyreview.info/articles/analysis/algorithmic-systems-consent-detail.
Goodell, G., and Tomaso A. (2019). A Decentralized Digital Identity Architecture. Frontiers in Blockchain. doi:
https://doi.org/10.3389/fbloc.2019.00017.
Greene, K.K., Kelsey, J., Frankli, J.M. (2016) Measuring the usability and security of permuted passwords on mobile
platforms. Technical report NISTIR 8040, National Institute of Standards and Technology, Information Access Division,
Information Technology Laboratory, 100 Bureau Drive (Mail Stop 8940) Gaithersburg, MD 20899-8940
Huhns, M. N., Stephens L. M. (2000) Multiagent Systems and Societies of Agents in Gerhard Weiss, Multiagent Systems: A
Modern Approach to Distributed Artificial Intelligence, MIT Press, USA
Jhaver, S., Karpfen, Y., & Antin, J. (2018). Algorithmic Anxiety and Coping Strategies of Airbnb Hosts. Proceedings of the
2018 CHI Conference on Human Factors in Computing Systems - CHI ’18, 1–12.
https://doi.org/10.1145/3173574.3173995
Jøsang, A. (2014). Identity management and trusted interaction in Internet and mobile computing. IET Information
Security. 8. 67-79. 10.1049/iet-ifs.2012.0133.
Latimer, P. and Duffy, M.. (2019). Deconstructing Digital Currency and Its Risks: Why ASIC Must Rise to the Regulatory
Challenge. Federal Law Review 47, no. 1 (March), pp. 121–50. https://doi.org/10.1177/0067205X18816237.
Marwick, A. E., and Boyd, D. (2014). Networked Privacy: How Teenagers Negotiate Context in Social Media. New Media &
Society 16, no. 7 (November 1), pp. 1051–67. https://doi.org/10.1177/1461444814543995.
McDonough, J. P. (1999). Designer Selves: Construction of Technologically Mediated Identity within Graphical, Multiuser
Virtual Environments. Journal of the American Society for Information Science, vol. 50, no. 10, pp. 855–69.
Monostori L., Váncza J., Kumara S.R.T., 2006, Agent-based systems for manufacturing, Annals of the CIRP, Vol. 55/2.
Nakamoto, S. (2008). Bitcoin: a peer-to-peer electronic cash system. Available at: https://bitcoin.org/bitcoin.pdf
Reid, Fergal, and Martin Harrigan (2012). An Analysis of Anonymity in the Bitcoin System. ArXiv:1107.4524 [Physics], May 7.
http://arxiv.org/abs/1107.4524.
Rouvroy, A., Berns, T. and Libbrecht, E.. (2013). “Algorithmic Governmentality and Prospects of Emancipation.” Réseaux No
177, no. 1 (October 14), pp. 163–96.
Oscar, S. “Competing Visions for AI.” Digital Culture & Society 4, no. 1 (March 1, 2018): 87–106.
https://doi.org/10.14361/dcs-2018-0107.
Tobin, A., and Reed, D. (2016). The Inevitable Rise of Self-Sovereign Identity. The Sovrin Foundation.
Voell, Z. (2020). 50 BTC Just Moved for First Time Since 2009 – But It Doesn’t Look Like Satoshi. Coindesk (May 20).
https://www.coindesk.com/50-btc-just-moved-for-first-time-since-2009-but-it-doesnt-look-like-satoshi
Winner, Langdon. (1978). Autonomous Technology. Technics-out-of-Control as a Theme in Political Thought. The MIT
Press.
Winner, L. (1980). Do Artifacts Have Politics? Daedalus, Vol. 109, No. 1, Modern Technology: Problem or Opportunity?
(Winter, 1980), pp. 121-136.
Wooldridge, M. (2000) Intelligent Agents in Gerhard Weiss, Multiagent Systems: A Modern Approach to Distributed
Artificial Intelligence, MIT Press, USA
Wooldridge, M., Jennings, N. R. (1995) Intelligent agents: Theory and practice. The Knowledge Engineering Review, 10(2)
Yermack, D. (2017) Corporate Governance and Blockchains. Review of Finance 21, no. 1 (March 1), pp. 7–31.
https://doi.org/10.1093/rof/rfw074.
122