European Conference on the Impact of Artificial Intelligence and Robotics

Activity Theory Analysis of RPA and Workforce in Financial Institutions

Denise Lakay, Nontobeko Mlambo — Tue, 22 Nov 2022 00:00:00 +0000

Financial institutions have been at the forefront of using Robotic Process Automation (RPA). Developing countries are moving towards using this emerging technology. Literature indicates various views on RPA and the workforce within financial institutions. This article, therefore, explores how RPA can be productively implemented in financial institutions. Activity Theory (AT) was applied to gain a deeper understanding of the challenges within financial institutions regarding the workforce employed. Using the six tenets of Activity Theory, this article looks at how these various areas impact the RPA adoption concerning the workforce. The use of RPA in financial institutions has assisted in processing mundane, repetitive tasks that do not require human intelligence. However, the AT tenets revealed the contradiction between RPA and the workforce. The challenges arise from a lack of understanding of how the two actors (RPA & workforce) can work in harmony and how both are reliant on one another. This paper uses qualitative methods to unpack the implications of RPA in financial institutions and the impact RPA has on the workforce. Various studies looked at the fear amongst the workforce regarding RPA, yet no empirical evidence exists to prove that RPA causes unemployment. This study demonstrates that communication is essential for introducing new technology.

Hey Siri: Exploring the Effect of Voice Humanity on Virtual Assistant Acceptance

Guillermo Calahorra Candao, María José Martín de Hoyos — Thu, 17 Nov 2022 00:00:00 +0000

This work in progress presents a proposed model to measure the acceptance of the use of voiced virtual assistants in commercial contexts. The use of these devices has increased after their implementation in smartphones and with the arrival of smart speakers in any household appliance. We propose the present model emphasizing the anthropomorphic aspect of the assistant to know if it has an emphasis on its acceptance. Unlike other models of acceptance, the contribution of the work is the inclusion of anthropomorphic variables such as human presence and human like voice. The work focuses on the virtual assistant Siri.

Value Co-Destruction in Digital Banking Transformation: Research Propositions

Darci De Borba — Thu, 17 Nov 2022 00:00:00 +0000

Digital Transformation imposes an invisible legacy on managers: the destruction of value. The ability of technology to disseminate services can lead to irreparable losses of value for companies, resulting in the decline of economic potential and imposing a dictatorship of gratuitousness. To research how this happens and propose solutions, I analyze the trend of value co-destruction (VCD) in the Digital Bank Transformation. The ability to understand and predict such changes is important to guide the processes of planning, implementing, and evaluating business decisions. Value creation is a central concept in the business literature, as companies create value through their operations and the delivery of services and products that meet the desires of their customers. However, value can also be destroyed, causing companies to fail and significant changes in the market. Through a semi-systematic literature review, I seek the theoretical guidelines of VCD in the context of online banking. We found 112 articles related to the topic and part of the systematic analysis of these articles is available in this work. The main objective of this theoretical essay is to highlight research propositions for the analysis of VCD in the context of Digital Banking Transformation. For the main delivery, it is necessary to: 1) delimit the concepts related to Digital Transformation and VCD; 2) understand how the VCD process is configured; 3) define the mechanisms related to the VCD; and 4) gather characteristics of the financial services segment in the context of Digital Transformation. Two proposals were proposed for future research.

Challenges and Opportunities- The Case for New Technologies’ Implementation in Romania

Decebal Dumitrescu, Martha Cristin Suciu, Mirela Aceleanu — Thu, 17 Nov 2022 00:00:00 +0000

Purpose: This article intends to identify, understand and explore the challenges and critical success factors (“CSF”) arising from implementation of new technologies in Romanian companies. We conduct a cross-industry partially explanatory, partially exploratory research focusing on 3 business cases whereby innovative technologies such as automation software, industrial robots and continuous steel melting through the use of electrical cauldron have been implemented.

Research Methodology/Design: We have used a case study approach. We’ve reviewed and mapped the existent academic literature on the subject matter, we have conducted interviews with business stakeholders, we have traced the identified challenges and CSF’ and measured their occurrence in relation to the case studies.

Findings:The results suggests that there are several pervasive challenges such as conflicts among the stakeholders, no management and/or employee buy-in, value misalignment and resistance to change that if appropriately mitigated can lead to successful outcomes. We’ve also identified a number recurring critical success factors that can ensure the effective implementation of the new technologies such as management buy in and understanding of value proposition.

Research limitations/implications:The current study further extends the existing research on challenges and CSF related to the implementation of technologies from the standpoint of innovation theory.

Practical Implications:The present article further expands on the exploration research performed so far on challenges and critical success factors related to the implementation of new technologies taking the scholarly research and validating the findings in the real business world.

Originality/value: The current research study provides practical evidence using cross-industry occurrence-based validation testing to indicate the challenges and critical success factors related to the implementation of new technologies in Romanian companies. It adds to the exploratory research on technology transformation as it pertains to jurisdictions that, more often than not, are outside the research radar.

An Integrative Literature Review on Leadership and Organizational Readiness for AI

Piper Frangos — Thu, 17 Nov 2022 00:00:00 +0000

Digital transformation is a reality for nearly all organizations today. Artificial intelligence (AI) plays a critical role in this transformation. The advent of machine learning and AI forces leaders to re-evaluate what it takes to lead an AI-driven firm successfully. As the scope and applicability of AI aggrandize, traditional leadership needs to evolve and meet the challenges and harness the opportunities this elusive technology presents. This literature review examines required leadership capabilities and organizational imperatives (beyond technology) for AI readiness and adoption. Based on literature, a new framework is presented to build on current leadership insights and technology adoption theories. The framework connects required leadership capabilities (agility, vision, engagement, ethics, and digital know-how) and organizational domains (knowledge, competence, and culture) to create a tool for executive leaders to drive AI adoption throughout their firms.

Assessment of Innovation Readiness and Technology Acceptance Using Immersive Sci-Fi Prototyping

Fiona Brunner, Thomas Keller, Elke Brucker-Kley — Thu, 17 Nov 2022 00:00:00 +0000

Digital transformation is ubiquitous and is generating unprecedented forms of innovation. However, it brings new challenges due to its comprehensive nature and potentially profound impacts. Technology assessment examines the long-term impacts of technologies on society and the environment. One aspect of this is a broad societal discourse, which is considered indispensable in a functioning democracy. To this end, various perspectives are sought, especially from experts in science, business, and politics, but also the opinion of the public. In practice, however, this poses a challenge. How is the public supposed to be able to form an informed opinion about a technological change, an upcoming innovation, if this change is multi-layered and not easily transparent and comprehensible? In this paper, attitude formation and attitude change are examined in more detail using a science fiction prototype in the democratic field of action. The main focus lies on immersion. The effects of immersive scenarios on attitudes toward technological innovations have not yet been sufficiently studied. The authors believe that conventional methods appeal mainly to cognition, but people are often driven by emotions. The immersive sci-fi prototyping method is designed to allow technological innovations to be experienced with virtual reality and thus also appeals to emotions in the process of forming attitudes. The authors hypothesize that with the immersive sci-fi prototyping method, a better starting point is provided to evaluate technology acceptance and innovation readiness. For this purpose, a laboratory experiment is conducted with a pre- and post-survey. The results of the low-immersion group, who click through the sci-fi prototype as hypertext in the browser, are compared with the high-immersion group, who experience the sci-fi prototype in a VR environment. The results show that both sci-fi prototyping and immersive sci-fi prototyping are suitable for capturing attitude. In the immersive sci-fi prototyping method, subjects are better able to visualise the technology and, in general, VR has shown a stronger impact on attitude change. These results relate to the digital democratic assistant presented in the sci-fi prototype.

The Automation of Computer Vision Applications for Real-time Combat Sports Video Analysis

Mr Evan Quinn, Dr Niall Corcoran — Thu, 17 Nov 2022 00:00:00 +0000

This study examines the potential applications of Human Action Recognition (HAR) in combat sports and aims to develop a prototype automation client that examines a video of a combat sports competition or training session and accurately classifies human movements. Computer Vision (CV) architectures that examine real-time video data streams are being investigated by integrating Deep Learning architectures into client-server systems for data storage and analysis using customised algorithms. The development of the automation client for training and deploying CV robots to watch and track specific chains of human actions is a central component of the project. Categorising specific chains of human actions allows for the comparison of multiple athletes' techniques as well as the identification of potential areas for improvement based on posture, accuracy, and other technical details, which can be used as an aid to improve athlete efficiency. The automation client will also be developed for the purpose of scoring, with a focus on the automation of the CV model to analyse and score a competition using a specific ruleset. The model will be validated by comparing performance and accuracy to that of combat sports experts. The primary research domains are CV, automation, robotics, combat sports, and decision science. Decision science is a set of quantitative techniques used to assist people to make decisions. The creation of a new automation client may contribute to the development of more efficient machine learning and CV applications in areas such as process efficiency, which improves user experience, workload management to reduce wait times, and run-time optimisation. This study found that real-time object detection and tracking can be combined with real-time pose estimation to generate performance statistics from a combat sports athlete's movements in a video.

Preparing Psychologists and Social Workers for the Daily Use of AI

Fredrik Åhs, Peter Mozelius, Majen Espvall — Thu, 17 Nov 2022 00:00:00 +0000

A daily use of Artificial Intelligence (AI) is becoming a fact in many fields today, and two of them are psychology and social work. At the same time as AI systems are used for predicting psychological treatments and for decisions in social welfare, higher education has few AI courses for these professions. Moreover, there are several examples in these fields where AI can make unethical decisions that need to be corrected by humans. To better understand the possibilities and challenges of AI in psychology and social work, professional users of AI services need a tailored education on how the underlying technology works. The aim of this paper is to present a project concept for the design and evaluation of a novel course in AI for professional development in psychology and social work. For the design and development of the course the guiding research question should be: What are the strengths and challenges with contemporary AI techniques regarding prediction, adaptivity and decision systems? The suggested AI course should be given as a technology enhanced online training to enable the idea of anytime and anywhere for full-time working participants. Course content and activities are divided into the four separate sections of: 1) The history of AI structured around the 'Three waves of AI', with o focus on the current third wave. 2) A section with a focus on AI techniques for prediction and adaptivity. Underlying techniques such as machine learning, neural networks, and deep learning will be conceptually described and discussed, but not on a detailed level. 3) An elaborated discussion on the relevance, usefulness and trust, and the at the difference between AI-based decision systems and AI-based decision support systems. 4) Finally, the fourth section should comprise the ethical aspects of AI, and discuss transparency and Explainable AI. An innovative approach of the project is to use a neuroscientific assessment of the education to understand how the education changes brain function relevant to evaluate AI based decision. This should be complemented with a qualitative evaluation based on semi-structured interviews.

Sentiment analysis for emotional navigation in written communication

Sofie Bergman, Niklas Humble — Thu, 17 Nov 2022 00:00:00 +0000

A subfield of artificial intelligence is that of natural language processing and sentiment analysis. The interest in sentiment analysis has increased with the development of the internet and web 2.0. With sentiment analysis it is possible to analyse the sentiment or emotions of written communication through dictionary-based sentiment analysis or machine learning algorithms. However, sentiment analysis also holds the potential of supporting people with disadvantages in interpreting the nuances in written communication. One such group is autistic people. The aim of the study is to examine autistic peoples’ perceptions of important design factors and functionality for an application with sentiment analysis to support emotional navigation in written communication.

This study has been conducted with the first steps of design science to outline the requirements of a potential application that can support autistic people to navigate the emotions in written communication with sentiment analysis. The problem to be addressed was identified through related research and one of authors’ own experience of navigating written communications with autism. The requirements for the application were the main focus for this study and has been the primarily concern for data collection. Data have been collected through semi-structured interviews with autistic people and analysed with thematic analysis.

Results of the study provide several important recommendations for the design of applications with sentiment analysis to support autistic people navigate the emotions in written communication. The study further provides an understanding of autistic peoples’ needs when navigating written communication. These findings can be used by researchers and developers to design support-applications with autistic peoples’ needs in the centre. An interesting next step of research would be to develop a prototype with the findings of this study addressed in the application’s functionality and design, which could then be evaluated on a larger scale.

A Framework for Auditing Multilevel Models using Explainability Methods

Debarati Bhaumik, Diptish Dey, Subhradeep Kayal — Thu, 17 Nov 2022 00:00:00 +0000

Multilevel models (MLMs) are increasingly deployed in industry across different functions. Applications usually result in binary classification within groups or hierarchies based on a set of input features. For transparent and ethical applications of such models, sound audit frameworks need to be developed. In this paper, an audit framework for technical assessment of regression MLMs is proposed. The focus is on three aspects: model, discrimination, and transparency & explainability. These aspects are subsequently divided into sub-aspects. Contributors, such as inter MLM-group fairness, feature contribution order, and aggregated feature contribution, are identified for each of these sub-aspects. To measure the performance of the contributors, the framework proposes a shortlist of KPIs, among others, intergroup individual fairness (Diff_{Ind_MLM}) across MLM-groups, probability unexplained (PUX) and percentage of incorrect feature signs (POIFS). A traffic light risk assessment method is furthermore coupled to these KPIs. For assessing transparency & explainability, different explainability methods (SHAP and LIME) are used, which are compared with a model intrinsic method using quantitative methods and machine learning modelling.

Using an open-source dataset, a model is trained and tested and the KPIs are computed. It is demonstrated that popular explainability methods, such as SHAP and LIME, underperform in accuracy when interpreting these models. They fail to predict the order of feature importance, the magnitudes, and occasionally even the nature of the feature contribution (negative versus positive contribution on the outcome). For other contributors, such as group fairness and their associated KPIs, similar analysis and calculations have been performed with the aim of adding profundity to the proposed audit framework. The framework is expected to assist regulatory bodies in performing conformity assessments of AI systems using multilevel binomial classification models at businesses. It will also benefit providers, users, and assessment bodies, as defined in the European Commission’s proposed Regulation on Artificial Intelligence, when deploying AI-systems such as MLMs, to be future-proof and aligned with the regulation.

Ethical and Legal Limits to the Diffusion of Self-Produced Autonomous Weapons

Elena Falletti, Chiara Gallese — Thu, 17 Nov 2022 00:00:00 +0000

The theme of self-produced weapons intertwines diversified ideas of an ethical, legal, engineering and data science nature. The critical starting point concerns the use of 3D printing for the self-production of weapons: the doctrinal and ethical discussion is open, while from a case-law point of view no published decisions have been found. From a technical point of view it should be noted that, being produced with materials other than metal, the weapons in question would increase their danger, since it would not be possible to ascertain their possession through metal detectors.

This possibility demonstrates how the combination of the application of 3D printing and AI can lead to further development of Autonomous Weapon Systems, especially drones, which are no longer confined to science fiction novels, but may appear on the market for goods and even become available for mass consumption, and it stresses the need for the promotion of negotiations for the drafting of an international treaty banning the production and use of lethal autonomous weapons.

The combination of such printers with biometric facial recognition algorithms raises concerns for the increasing issues of physical, individual and collective safety that may arise. In fact, the biometric recognition technology allows the identification of individuals through the measurement and analysis of the somatic or behavioural traits; it is based on intelligent software, modelled on the human ability to recognize and identify faces by collecting and analysing huge amounts of data, and it is able to evolve its skills beyond its programmer’s initial intention. It is clear that allowing self-production of such devices by non-expert users could produce more damages than benefits.

The purpose of this contribution is to study how to regulate the effects of such self-made autonomous robots, since their use may have a devastating and disruptive effect on public integrity and social peace, especially in case of violent riots.

Suggestions for a Revision of the European Smart Robot Liability Regime

Chiara Gallese — Thu, 17 Nov 2022 00:00:00 +0000

In recent years, the need for regulation of robots and Artificial Intelligence, together with the urgency of reshaping the civil liability framework, has become apparent in Europe. Although the matter of civil liability has been the subject of many studies and resolutions, multiple attempts to harmonize EU tort law have been unsuccessful so far, and only the liability of producers for defective products has been harmonized so far. In 2021, by publishing the AI Act proposal, the European Commission reached the goal to regulate AI at the European level, classifying smart robots as ”high-risk systems”. This new piece of legislation, albeit tackling important issues, does not focus on liability rules. However, regulating the responsibility of developers and manufacturers of robots and AI systems, in order to avoid a fragmented legal framework across the EU and an uneven application of liability rules in each Member State, is still an important issue that raises many concerns in the industry sector. In particular, deep learning techniques need to be carefully regulated, as they challenge the traditional liability paradigm: it is often not possible to know the reason behind the output given by those models, and neither the programmer nor the manufacturer is able to predict the AI behavior. For this reason, some authors have argued that we need to take liability away from producers and programmers when robots are capable of acting autonomously from their original design, while others have proposed a strict liability regime. This article explores liability issues about AI and robots with regards to users, producers, and programmers, especially when the use of machine learning techniques is involved, and suggests some regulatory solutions for European lawmakers.

ENSIGHTS: Intelligent Monitoring of Electric Power Transmission Assets

Thu, 17 Nov 2022 00:00:00 +0000

This work aims to use Data Science techniques to build predictive models that will eventually improve maintenance plans regarding power transformers by reducing shutoffs and transmission downtime. This work is part of a 36-month long Research and Development (R&D) project started on January 2021, as of the writing of this report the project is halfway through. The analytic models described herein have already been tested while most of the remaining work is yet to be done. In a single Data Lake environment, we will consolidate several databases from information systems that support the company's operation and maintenance processes. Various machine learning models will run on this data, and their results will appear in a dashboard, alongside several traditional indicators. This process will be integrated and consolidated in a computing platform with cloud architecture. We use Machine Learning (ML) to develop the models. Based on an Agnostic Probably Approximately Correct (PAC) Learning study of the available datasets estimating their Vapnik–Chervonenkis dimension, we choose Random Forests (RF) algorithms to be used in the new indicators. So far, the project has produced two new indicators: Chromatographic Assay Indicator (CAI) and Electrical Failure Risk Indicator (EFRI). The CAI indicator evaluation uses a Random Forest Algorithm trained with an external dataset due to the small number of power transformers failures in the O&M data. This indicator performed much better than classical chromatographic indicators to predict electric or temperature problems on the test set. The EFRI indicator correlates monitoring data available from an existing Supervisory Control and Data Acquisition (SCADA) system with maintenance data from an existing Enterprise Resource Planning (ERP) system through a RF algorithm capable of alerting to a higher risk of electric failure. ANEEL funds this work as R&D project PD-00394-1907/2019 titled “Aplicabilidade de nova tecnologia voltada para o desenvolvimento de um modelo de monitoramento inteligente dos ativos de transmissão”.

An Outlook of Digital Twins in Offensive Military Cyber Operations

Clara Maathuis — Thu, 17 Nov 2022 00:00:00 +0000

The outlook of military cyber operations is changing due to the prospects of data generation and accessibility, continuous technological advancements and their (public) availability, technological and human (inter)connections increase, plus the dynamism, needs, diverse nature, perspectives, and skills of experts involved in their planning, execution, and assessment phases respecting (inter)national aims, demands, and trends. Such operations are daily conducted and recently empowered by AI to reach or protect their targets and deal with the unintended effects produced through their engagement on them and/or collateral entities. However, these operations are governed and surrounded by different uncertainty levels e.g., intended effects prediction, consideration of effective alternatives, and understanding new dimensions of possible (strategic) future(s). Hence, the legality and ethicality of such operations should be assured; particularly, in Offensive Military Cyber Operations (OMCO), the agents involved in their design/deployment should consider, develop, and propose proper (intelligent) measures/methods. Such mechanisms can be built embedding intelligent techniques based on hardware, software, and communication data plus expert-knowledge through novel systems like digital twins. While digital twins find themselves in their infancy in military, cyber, and AI academic research and discourses, they started to show their modelling and simulation potential and effective real-time decision support in different industry applications. Nevertheless, this research aims to (i) understand what digital twins mean in OMCO context while embedding explainable AI and responsible AI perspectives, and (ii) capture challenges and benefits of their development. Accordingly, a multidisciplinary stance is considered through extensive review in the domains involved packaged in a design framework meant to assist the agents involved in their development and deployment.

Narratives That Speak AI Lingua? AI Vocabulary in Listed Companies’ Annual Reports

Dagmar Monett, Claudia Lemke , Liadan Anandarajah, Tom Brandherm — Thu, 17 Nov 2022 00:00:00 +0000

Narratives about intelligent artefacts have influenced both the public’s imaginary and the actual development of the AI field since its foundation. Yet, in times where the field seems to be flourishing on the one hand, but rushing into an AI winter on the other, factual narratives about AI applications and advancements are more essential than ever. What is the gap between the actual capabilities of today’s AI and the vocabulary used to report about them? In particular, what is the AI lingua used in official, legal documents in business? To find out, we analysed leading share index companies’ annual reports from a representative fraction of the German economy (DAX 30), as a starting step in this direction. In this paper, we present a fact-based methodology for systematically assessing the true state of enterprise AI of those companies. Our initial empirical investigation covers only the annual reports of leading listed German enterprises in the DAX 30 as of May 2021 (i.e. before the DAX’s expansion to 40 members). For this concrete example, we collected their annual reports from 2010 to 2020 (N=312). We then built upon previous work by extending natural language processing (NLP) algorithms we developed for these purposes. The idea is to systematically process and automatically detect the use of AI-related terminology in those annual reports. Such a terminology is part of a classification schema we introduce for differentiating concrete types of AI-related terms. We also compare different NLP libraries regarding their suitability and speculate on the reasons behind the poor performance of some of them. Furthermore, we look at relevant AI keywords and phrases, thereby conducting a human-based semantic analysis of the context – tasks that machines still cannot do effectively. We also give guidance on how to proceed in similar studies, i.e. on how to extend our methodology and the key findings to other national economies. This way, we are contributing not only to an informed perception about the state of enterprise AI, but also to filling the gap between the narratives it uses and the actual state of AI development.

Would you Like to Have Your Social Skills Assessed by a Softbot? AI-Supported Recruitment Processes

Peter Mozelius, Amir Jama, Aile Castberg — Thu, 17 Nov 2022 00:00:00 +0000

In parallel with the increased use of Artificial Intelligence (AI) in recruitment processes, there is also an ongoing
discussion on the dehumanisation in automated recruitment. On one hand AI-based recruitment has the potential to
reduce human bias, on the other hand there are parts of the process that still need human judgement. Another concern is
that the identified dehumanisation could harm the relationship between employees and employers. Research indicates
that AI-based technologies definitely have the potential to increase the efficiency of the recruitment process by replacing
humans in time-consuming tasks. Less research has been conducted on the human perceptions about AI-based
recruitment. In a time when AI-based recruitment tools are used in a rapidly increasing number of companies and
organisations, it is important to better explore the human side of the process. Therefore, this paper investigates: What
are the perceptions of the job candidate conditions in automatised and AI-based recruitment processes? This study was
conducted with a qualitative approach with data gathered from candidates and recruiters that all had experiences from
AI-based recruitment processes. Four candidates and two recruiters were chosen with the idea of a purposive sampling.
Answers from six audio recorded semi-structured interviews were categorised in a deductive thematic analysis. The
theoretic lens for the study was the Model of Applicant Reaction to Selection. Findings showed that the informants had a
negative attitude towards the dehumanised recruitment process. The most obvious finding was the general critique
towards the AI-based assessment of candidates' social skills. At the same time, the majority of the informants agreed that
AI-based recruitment tools have the potential to make time-consuming administrative tasks more efficient. Only one
informant was willing to go through a completely AI-based recruitment process, and all informants pointed out different
ways in which the recruitment tools need to be improved. The conclusion is that the AI-based recruitment tools must be
made more transparent and used as a support for decision-making rather than being the decision maker. The
recommendation is a hybrid solution, where AI-based tools are used to assist and create the basis for well-informed
human decisions.

Ethics and Accountability of Care Robots

Jyri Rajamäki, Jaakko Helin — Thu, 17 Nov 2022 00:00:00 +0000

The subject of this paper is ethical and responsibility issues relating to the development and acquisition of robotics in healthcare. The purpose of the paper is to study previous scientific publications and research related to the topic and to clarify which questions, aspects, and concerns are most relevant when considering ethics and responsibility issues related to care robots. In the second phase, ideas from different stakeholders regarding the viewpoints are studied, and those ideas are compared to the ones presented in previous publications. The aim of this study is to find solutions to the issues presented in scientific literature and, also, to find new issues for consideration and further studies. The study is qualitative, and a theme interview was utilized as the main method for acquiring knowledge. The study is a part of the SHAPES Horizon 2020 project. From the perspective of SHAPES, the aim of the study is to provide useful knowledge for the project, which would in part promote the goal of SHAPES, i.e., the development of an international healthcare ecosystem. Based on the results of the study, it can be argued that the issues presented in previous academic publications regarding the ethics and accountability of robots in practical healthcare work are not relevant. Both the legislation and the logic of the AI algorithms used by care robots prevent those situations presented in previous academic discussions in which robots would presumably be forced to make decisions demanding ethical consideration. The results also point toward the fact that current legislation does not limit the development of healthcare robots more than it limits healthcare work in general. Thus, the considerations of ethics regarding care robots should rather be focused on the threshold values used by robots, when making interpretations, as well as the data used for the purpose of machine learning. These were identified as potential subjects for further research.

Dark Patterns and Epistemic Ignorance: an Educational Crisis

Stockman, Richard Wilson — Thu, 17 Nov 2022 00:00:00 +0000

In this paper, we discuss how online AI stimulates epistemic ignorance. Early visions of online information search and retrieval processes proposed a utopian and empowering space for individuals. Today’s crisis paradoxically presents us with an unprecedented accumulation of new information and access to it, yet also the colonisation of this knowledge by those who seek to erode critical thought. By ‘epistemic ignorance’, we mean the condition which is systematically created by the patterns of mis- and disinformation that prevent knowledge seekers from gaining verified knowledge. We argue not only has the ‘knower’ or knowledge seeker become the ‘known’ (sometimes without knowing it), their ability ‘to know’ is also intentionally manipulated by dark patterns. Moreover, their ‘known’ status allows for their subtle indoctrination, and erosion of criticality. This makes the crisis an educational one. To illustrate, we consider epistemic mechanisms on Facebook pertaining to the early stage of the Covid-19 pandemic. We contend these ‘dark AI patterns’ intentionally aim for systemic indoctrination, and affective indoctrination, by engaging in the construction of epistemic ignorance. Our focus is on the political agenda; which is common in the wider discussion of indoctrination in education. Many educational philosophers have taken a critical interest in the power of education to indoctrinate. The formal educational space is an effective vehicle to do so – and now the informal education we receive through social media is as well. Through algorithms, we are taught to think a certain way. This new crisis has not yet been considered an educational one, while in every moment, the coercive powers of online AI drive audiences towards greater uncritical acceptance of knowledge and information. Perhaps we can reverse the educational oppression with the introduction of ‘light patterns’.

A Roadmap to Artificial Intelligence: Navigating Core Impacts to Successfully Transform Organisations

Stephen Treacy — Thu, 17 Nov 2022 00:00:00 +0000

Artificial Intelligence (AI) is a highly disruptive technology that will have major effects on the business world over the coming years. It has the potential to allow companies to achieve major efficiency gains and a more productive workforce through automating existing processes, providing deeper levels of analytics, providing better customer support, and increasing security. On the other hand, it may lead to lower staff levels and a drop in existing employee morale. Given the complexities of these projects, AI will only benefit organisations if they understand its capabilities in addition to its shortcomings. This investigation addresses the predicted impact on skills, roles and employee morale of artificial intelligence on the workforce of the future as AI continues to become more prevalent in our society.

We investigate these impacts of AI specifically across four key industries by engaging in interviews with experts in the field to answer two research questions: (i) What are the core impacts of introducing AI systems in the workplace?, and; (ii) How can organisations develop AI projects for successful transformation? The inclusion strategy for this research were professionals who were highly knowledgeable in the area, and from our findings we were able to identify several impacts that AI made to companies developing these projects; namely employment levels, workforce morale, and process efficiency. With these insights, we subsequently developed a roadmap which contains the recommended steps and decisions that are necessary for successfully introducing AI to an organisation. This roadmap visualises the key decisions and steps that are critical for any AI based initiative for organisations, which will provide practitioners with a higher level of understanding of what is expected, in addition to enabling more effective collaboration with the system developers. Furthermore, this roadmap allows organisations to take a positive and proactive approach to designing these systems with their workforce in mind and to prepares them for the implications with the development, deployment, and use of these AI systems.

Action Rule Mining With Predications on Semantic Representations of Data

Li-Shiang Tsay — Thu, 17 Nov 2022 00:00:00 +0000

Without a doubt, the World Wide Web (WWW) has already altered the way we share knowledge. The exploitation of the web is one of the biggest challenges in the area of intelligent information management. The Semantic Web was presented to ease this situation by taking the WWW into a distributed global system for knowledge representation and computing. In this work, we study action rule mining in the semantic web data and aim to assist users in uncovering previously unknown and potentially useful workable strategies. We present rule-based action rules and object-based action rules for extracting actionable patterns from a large graph dataset without the extra step of converting graph data into transaction data.

Politicians and Artificial Intelligence Refusal: Brief Considerations

Marius Vacarelu — Thu, 17 Nov 2022 00:00:00 +0000

Technological development of the last decades changed not only working procedures but also economic hierarchies and – in many cases – it offered an instrument for ambitious leaders to bring a new perspective to countries, continents, or even the world in different areas: economy, law, politics, etc.

Inside these new dimensions of human life, politicians are asked to create rules for societies, but also their specific tasks. At the same time, politicians are asked to think about the future and to settle the main directions for national development, in such a way to profit for today and the next generations. For such development plans, politicians need to consult many data and create a specific legal framework, able to increase people's skills in every area, in a coherent vision that includes Artificial Intelligence.

Artificial Intelligence might have a special field of action in political competition, which must be regulated by the same actors able to use it. Such a specific legal possibility – to be able to regulate one of your tools – offers to politicians many interrogations about the limits of Artificial Intelligence use. Politics is a matter of power and history shows that many times politicians use many tools and administrative procedures to preserve or achieve power. Artificial Intelligence could be used for the same purposes and both scholars and citizens will be important to study how politicians will regulate it. In this case, an important topic is the Artificial Intelligence acceptance in political completions and if politicians will try to regulate its use with a specific interdiction on the political area. It will be also important to see if such interdictions will consider the danger for democratic institutions or other reasons like psychological dangers to the human mind, costs of implementation, etc.

A Conceptual Decision Tree for the Automation of Architecture Applied to Housing

Francine van Tonder, Pierre Brink — Thu, 17 Nov 2022 00:00:00 +0000

In 1994, the Nelson Mandela government identified the lack of housing as the most severe social problem faced by South Africa. However, housing information is often inaccessible to those affected by inadequate low-cost housing. The right information at the right time has the potential to not only address the inadequacy of low-cost housing but also to address, in part, other social ills such as unemployment, poverty and government corruption, as well as larger issues such as the climate emergency. This paper presents a conceptual decision tree to govern the knowledge management of architectural information for the possible automation of architecture. Knowledge management, big data and machine learning are the precursors of artificial intelligence, a technology that could further aid in addressing the inadequacy of low-cost housing. A decision tree is the first step. For this paper, the information within frames forms the nodes on the conceptual decision tree. This decision tree is presented graphically and tested hypothetically on a low-cost housing unit. The research findings indicate that there is a noteworthy overlap in architectural information of low-cost housing information presented thereby validating the possible benefit of architectural information knowledge management.