AI has become one of the most powerful drivers of social change, transforming economies, impacting politics and wars, and reshaping how citizens live and interact. Nevertheless, the implementation of AI can have adverse effects on peoples’ lives. This dissertation first examines the relationship between artificial intelligence and public law, mainly in two domains, administrative law and criminal law. It also provides a clear insight into the potential impact of AI applications on fundamental rights in the legal context of the European Union. Four selected fundamental rights, Human Dignity, Data Protection and Right to Privacy, Equality and Non-discrimination, and Access to Justice, are examined. The dissertation further explores the European Commission's new proposed AI regulation, which was proposed in April 2021. The proposal aims to put forward a risk- based approach for a harmonized EU legislation by considering the ethical and human sides and without unnecessarily restricting the development of AI technologies. The study focuses on examples from Sweden throughout the study and lastly, examines the Swedish approach in the context of the principles of good administration.

This article aims towards developing further understanding regarding the relation between the media phenomenon podcasts and source credibility amongst young adults in Sweden. Furthermore, the foundation for this essay is based on the previous and current empiricism regarding both podcasts and source credibility. This study has evaluated and discussed the function that podcasts contribute with in regards to bridging the gap between social media and journalistic publishing. To answer this study's aim of purpose, a qualitative and a quantitative research method have been applied. These research methods have consisted of a digital survey and a focus group survey. The purpose behind the research was to examine how young adults living in Sweden experience and reflect on the subject of podcasts in correlation with source credibility. The research was narrowed down to a specific set of three resource categories amongst podcasts. It was concluded from these studies’ research methods that similarities and differences were present in relation to already existing empiricism. Furthermore was the extubated data applied to the Parasocial Interaction Theory and Uses and Gratification Theory.  The extubated data showed how young adults consider themselves source critical in relation to information produced through podcasts. Furthermore it was revealed that young adults' source credibility is highly based on who the publisher and podcast host is and acts. After applying the two theories to the collected data it was illustrated that young adults’ source critical thinking was affected by their perceived personal relations towards the podcast hosts. It became apparent that podcasts within the public service sector were perceived more credible regardless of the podcast subject and podcast host. In further addition the collected data presented that credibility and source criticism amongst young adults is partly based on which genre of podcast that is consumed. This essay also conducts a discussion regarding how the relation between podcasts and source credibility can develop further in the future. / Följande arbete ämnar att bilda en ökad förståelse kring mediefenomet podcasts i relation till det samhällsaktuella begreppet källkritik. Således redovisas det hur tidigare forskning tillhörande forskningsområdena har bidragit till den nuvarande empirin inom ämnet. Uppsatsen beskriver och diskuterar podcasts som en medial kanal som verkar mellan sociala medier och journalistiskt publicerade verk. För uppsatsens syfte har en kvalitativ och en kvantitativ studie genomförts. Dessa forskningsmetoder utformades som en digital enkät och en fokusgruppsundersökning. Arbetet utgick från att undersöka hur unga vuxna i Sverige upplever och reflekterar kring podcasts i relation till källkritik inom tre olika podcastgenrer. Genom arbetets forskningsmetoder synliggjordes likheter och olikheter i relation till tidigare forskning och applicerades även i relation till två olika teorier. Dessa två teorier var Parasocial Interaction Theory och Uses and Gratification Theory. Studien visade att unga vuxna är generellt källkritiska till det material de får ta del av via podcasts, men att trovärdigheten och nivån av källkritik avgörs beroende på vem avsändaren och podvärden är. Genom applicering av de båda teorierna till studiens data från undersökningarna, synliggjordes det att lyssnarnas uppfattade personliga relationer till podcastvärdarna påverkade deras källkritiska tänkande. Ytterligare framgick det att podcasts med public service som avsändare ansågs vara mer trovärdiga oberoende ämnet och podcastvärd. Studien visade även att trovärdigheten och källkritiken hos unga vuxna är delvis beroende på vilken kategori av podcast som konsumeras. Uppsatsen för även en diskussion om hur källkritik kan appliceras i relation till mediefenomet podcasts i framtiden.

Artificial intelligence and machine learning has become an important part of society and today businesses compete in a new digital environment. However, scholars and regulators are concerned with these technologies' societal impact as their use does not come without risks, such as those stemming from transparency and accountability issues. The potential wrongdoing of these technologies has led to guidelines and future regulations on how they can be used in a trustworthy way. However, these guidelines are argued to lack practicality and they have sparked concern that they will hamper organisations' digital pursuit for innovation and competitiveness. This master’s thesis aims to contribute to this field by studying how teams can work with risk mitigation of risks associated with machine learning. The scope was set on capturing insights on the perception of employees, on what they consider to be important and challenging with machine learning risk mitigation, and then put it in relation to research to develop practical recommendations. The master’s thesis specifically focused on the financial technology sector and the use of machine learning in credit scoring. To achieve the aim, a qualitative single case study was conducted. The master’s thesis found that a combination of processes and capabilities are perceived as important in this work. Moreover, current barriers are also found in the single case. The findings indicate that strong responsiveness is important, and this is achieved in the single case by having separation of responsibilities and strong team autonomy. Moreover, standardisation is argued to be needed for higher control, but that it should be implemented in a way that allows for flexibility. Furthermore, monitoring and validation are important processes for mitigating machine learning risks. Additionally, the capability of extracting as much information from data as possible is an essential component in daily work, both in order to create value but also to mitigate risks. One barrier in this work is that the needed knowledge takes time to develop and that knowledge transferring is sometimes restricted by resource allocation. However, knowledge transfer is argued to be important for long term sustainability. Organisational culture and societal awareness are also indicated to play a role in machine learning risk mitigations. / Artificiell intelligens och maskininlärning har blivit en betydelsefull del av samhället och idag konkurrerar organisationer i en ny digital miljö. Forskare och regulatorer är däremot bekymrade gällande den samhällspåverkan som sådan teknik har eftersom användningen av dem inte kommer utan risker, såsom exempelvis risker som uppkommer från brister i transparens och ansvarighet. Det potentiella olämpliga användandet av dessa tekniker har resulterat i riktlinjer samt framtida föreskrifter på hur de kan användas på ett förtroendefullt och etiskt sätt. Däremot så anses dessa riktlinjer sakna praktisk tillämpning och de har väckt oro då de möjligen kan hindra organisationers digitala strävan efter innovation och konkurrenskraft. Denna masteruppsats syftar till att bidra till detta område genom att studera hur team kan arbeta med riskreducering av risker kopplade till maskininlärning. Uppsatsens omfång lades på att fånga insikter på medarbetares uppfattning, för att sedan ställa dessa i relation till forskning och utveckla praktiska rekommendationer. Denna masteruppsats fokuserade specifikt på finansteknologisektorn och användandet av maskininlärning inom kreditvärdering. En kvalitativ singelfallstudie genomfördes för att uppnå detta mål. Masteruppsatsen fann att en kombination av processer och förmågor uppfattas som viktiga inom detta arbete. Dessutom fann fallstudien några barriärer. Resultaten indikerar att en stark förmåga att reagera är essentiellt och att detta uppnås i fallstudien genom att ha tydlig ansvarsfördelning och att teamen har stark autonomi. Vidare så anses standardisering behövas för en högre nivå av kontroll, samtidigt som det bör vara implementerat på ett sådant sätt som möjliggör flexibilitet. Fortsättningsvis anses monitorering och validering vara viktiga processer för att mitigera maskininlärningsrisker. Dessutom är förmågan att extrahera så mycket information från data som möjligt en väsentlig komponent i det dagliga arbetet, både för värdeskapande och för att minska risker. En barriär inom detta arbetet är att det tar tid för den behövda kunskapen att utvecklas och att kunskapsöverföring ibland hindras av resursallokering. Kunskapsöverföring anses däremot vara viktigt för långsiktig hållbarhet. Organisationskultur och samhällsmedvetenhet indikeras också påverka minskningen av risker kring maskininlärning.

<p dir="ltr">Despite the recent advancement and exponential growth in the utility of deep learning models across various fields and tasks, we are confronted with emerging challenges. Among them, one prevalent issue is the biases inherent in deep models, which often mimic stereotypical or subjective behavior observed in data, potentially resulting in negative societal impact or disadvantaging certain subpopulations based on race, gender, etc. This dissertation addresses the critical problem of fairness and bias in machine learning from diverse perspectives, encompassing both data biases and model biases.</p><p dir="ltr">First, we study the multifaceted nature of data biases to comprehensively address the challenges. Specifically, the proposed approaches include the development of a generative model for balancing data distribution with counterfactual samples to address data skewness. In addition, we introduce a novel feature selection method aimed at eliminating sensitive-relevant features that could potentially convey sensitive information, e.g., race, considering the interrelationship between features. Moreover, we present a scalable thresholding method to appropriately binarize model outputs or regression data considering fairness constraints for fairer decision-making, extending fairness beyond categorical data.</p><p dir="ltr">However, addressing fairness problem solely by correcting data bias often encounters several challenges. Particularly, establishing fairness-curated data demands substantial resources and may be restricted by regal constraints, while explicitly identifying the biases is non-trivial due to their intertwined nature. Further, it is important to recognize that models may interpret data differently by their architectures or downstream tasks. In response, we propose a line of methods to address model bias, on top of addressing the data bias mentioned above, by learning fair latent representations. These methods include fair disentanglement learning, which projects latent subspace independent of sensitive information by employing conditional mutual information, and a debiased contrastive learning method for fair self-supervised learning without sensitive attribute annotations. Lastly, we introduce a novel approach to debias the multimodal embedding of pretrained vision-language models (VLMs) without requiring downstream annotated datasets, retraining, or fine-tuning of the large model considering the constrained resource of research labs.</p>

The goal of the diploma thesis is to put forward a procedure for implementation of an institutional repository at the Czech University of Life Sciences in Prague with the aid of an analysis on the current situation of institutional repositories in Czechia. The analysed repositories are selected using OpenDOAR and ROAR registries. The first part of the thesis describes the software used to implement the institutional repositories in Czechia. In the next part tools, standards and certifications relevant for trustworthy institutional repositories such as DRAMBORA and PLATTER, standards ISO 16363 and ISO 14721 and the so called lesser certifications Data Seal of Approval and Nestor Seal of Trustworthy Digital Archives are described. Following part focuses on policies and operation of those analysed repositories. The last part of the thesis deals with a description of current practice on the Czech University of Life Sciences, the conducted quantitative analysis among the researchers of the university and the aforementioned procedure for implementation of an institutional repository using the tool PLATTER. The conclusion summarizes the issues of institutional repositories in Czechia.

In the context of designing and implementing ethical Artificial Intelligence (AI), varying perspectives exist regarding developing trustworthy AI for autonomous cars. This study sheds light on the differences in perspectives and provides recommendations to minimize such divergences. By exploring the diverse viewpoints, we identify key factors contributing to the differences and propose strategies to bridge the gaps. This study goes beyond the trolley problem to visualize the complex challenges of trustworthy and ethical AI. Three pillars of trustworthy AI have been defined: transparency, reliability, and safety. This research contributes to the field of trustworthy AI for autonomous cars, providing practical recommendations to enhance the development of AI systems that prioritize both technological advancement and ethical principles.

No description available.

The rapid advancement of AI technology emphasizes the importance of developing practical and ethical frameworks to guide its evolution and deployment in a responsible manner. In light of more complex AI and its capacity to influence society, AI researchers and other prominent individuals are now indicating that AI evolution has to be regulated to a greater extent. This study examines the practical implementation of Responsible AI guidelines in an organization by investigating the challenges encountered and proposing solutions to overcome them. Previous research has primarily focused on conceptualizing Responsible AI guidelines, resulting in a tremendous number of abstract and high-level recommendations. However, there is an emerging demand to shift the focus toward studying the practical implementation of these. This study addresses the research question: ‘How can an organization overcome challenges that may arise when implementing Responsible AI guidelines in practice?’. The study utilizes the guidelines produced by the European Commission’s High-Level Expert Group on AI as a reference point, considering their influence on shaping future AI policy and regulation in the EU. The study is conducted in collaboration with the telecommunications company Ericsson, which henceforth will be referred to as 'the case organization’, which possesses a large global workforce and headquarters in Sweden. Specific focus is delineated to the department that works on developing AI internally for other units with the purpose of simplifying operations and processes, which henceforth in this study will be referred to as 'the AI unit'. Through an inductive interpretive approach, data from 16 semi-structured interviews and organization-specific documents were analyzed through a thematic analysis. The findings reveal challenges related to (1) understanding and defining Responsible AI, (2) technical conditions and complexity, (3) organizational structures and barriers, as well as (4) inconsistent and overlooked ethics. Proposed solutions include (1) education and awareness, (2) integration and implementation, (3) governance and accountability, and (4) alignment and values. The findings contribute to a deeper understanding of Responsible AI implementation and offer practical recommendations for organizations navigating the rapidly evolving landscape of AI technology.

Trust in information systems stem from two key properties of responses to queries regarding the state of the system, viz., i) authoritativeness, and ii) unbiasedness. That the response is authoritative implies that i) the provider (source) of the response, and ii) the chain of delegations through which the provider obtained the authority to respond, can be verified. The property of unbiasedness implies that no system data relevant to the query is deliberately or accidentally suppressed. The need for guaranteeing these two important properties stem from the impracticality for the verifier to exhaustively verify the correctness of every system process, and the integrity of the platform on which system processes are executed. For instance, the integrity of a process may be jeopardized by i) bugs (attacks) in computing hardware like Random Access Memory (RAM), input/output channels (I/O), and Central Processing Unit( CPU), ii) exploitable defects in an operating system, iii) logical bugs in program implementation, and iv) a wide range of other embedded malfunctions, among others. A first step in ensuing AU properties of geographic queries is the need to ensure AU responses to a specific type of geographic query, viz., point-location. The focus of this dissertation is on strategies to leverage assured point-location, for i) ensuring authoritativeness and unbiasedness (AU) of responses to a wide range of geographic queries; and ii) useful applications like Secure Queryable Dynamic Maps (SQDM) and trustworthy redistricting protocol. The specific strategies used for guaranteeing AU properties of geographic services include i) use of novel Merkle-hash tree- based data structures, and ii) blockchain networks to guarantee the integrity of the processes.

Local Law Enforcement and Immigration: Lessons and Recommendations from Police Executives (2007-2021) Tonya D. Chapman ABSTRACT The Immigration and Reform Control Act (IRCA) of 1986 and the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (IIRIRA) authorized the U.S. Immigration and Customs Enforcement agency (ICE) to enter into memoranda of agreement with local law enforcement under section 287(g). The 287(g) program includes the Task Force Model (TFM), Jail Enforcement Model (JEM), Secure Communities (SC), the Priority Enforcement Program (PEP) model and the Warrant Service Officer (WSO) Model, which authorizes specific responsibilities of immigration enforcement to local law enforcement agencies. This dissertation examines the impact of local law enforcement's participation in the various 287(g) programs from the perspective of law enforcement executives. Local law enforcement was granted the responsibility in part because Congress and local elected officials believed that immigration increased crime. However, as of 2022, little research on the nexus between crime and immigration supports that claim; nor does it support the claim that crime rates fell as a result of local law enforcement's participation in the 287(g) programs. Consistent with prior research, this dissertation finds that immigration enforcement has a "null or non-significant" effect on crime in these jurisdictions in comparison to jurisdictions that did not participate in the 287(g) programs. Moreover, this dissertation shows that law enforcement's participation in immigration enforcement led to unintended consequences, including adverse impacts on police legitimacy (trust and fear), perceived crime reporting by immigrant communities, and their community policing efforts. This research provides guidance on best practices to law enforcement in an effort to re-imagine the profession in accordance with procedural justice principles. It examines whether and how immigration enforcement has posed challenges for building trust, legitimacy, community engagement and transparency for law enforcement; looks at whether federal mandates and immigration enforcement affected the advancement of community policing and procedural justice; provides insight on lessons learned from law enforcement's perspective; and contributes to research on the immigration-crime nexus. / Doctor of Philosophy / Local Law Enforcement and Immigration: Lessons and Recommendations from Police Executives (2007-2021) Tonya D. Chapman GENERAL AUDIENCE ABSTRACT Section 287(g) under the Immigration and Reform Control Act (IRCA) of 1986 and the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (IIRIRA) authorized the U.S. Immigration and Customs Enforcement agency (ICE) to enter into memoranda of agreement with local law enforcement agencies to participate in immigration enforcement. Under Section 287(g), ICE implemented 5 programs, including the Task Force Model (TFM), Jail Enforcement Model (JEM), Secure Communities (SC), the Priority Enforcement Program (PEP) model, and the Warrant Service Officer (WSO) model. This dissertation examines the impact of local law enforcement's participation in the 287(g) programs, from the perspective of law enforcement executives. This dissertation shows that law enforcement's participation in immigration enforcement led to unintended consequences, including adverse impacts on police legitimacy (trust and fear), perceived crime reporting by immigrant communities, and their community policing efforts. The dissertation also finds that immigration enforcement has a "null or non-significant" effect on crime. This research provides guidance on best practices to law enforcement in an effort to re-imagine the profession in accordance with fair and impartial policing principles.