Magnus Westerlund

Dennis Vetter, Jesmin Jahan Tithi, Magnus Westerlund et al.

Assessing the trustworthiness of artificial intelligence systems requires knowledge from many different disciplines. These disciplines do not necessarily share concepts between them and might use words with different meanings, or even use the same words differently. Additionally, experts from different disciplines might not be aware of specialized terms readily used in other disciplines. Therefore, a core challenge of the assessment process is to identify when experts from different disciplines talk about the same problem but use different terminologies. In other words, the problem is to group problem descriptions (a.k.a. issues) with the same semantic meaning but described using slightly different terminologies. In this work, we show how we employed recent advances in natural language processing, namely sentence embeddings and semantic textual similarity, to support this identification process and to bridge communication gaps in interdisciplinary teams of experts assessing the trustworthiness of an artificial intelligence system used in healthcare.

Valerie Bürger, Marlie Besouw, Jana Fehr et al.

Meta-research and Trustworthy AI (TAI) share common goals, namely improving evidence, robustness, and transparency, yet there is very little interplay between the two fields. To investigate the potential benefits of closer collaboration between the domains of TAI in healthcare and meta-research, we convened an interdisciplinary workshop funded by the Volkswagen Foundation in February 2025. The workshop aimed to collaboratively examine key tensions in translating AI ethics principles into practice and to identify potential solutions informed by meta-research approaches. A Design Thinking-informed co-creation approach was followed by an inductive descriptive analysis of the outputs. Our results demonstrate how meta-research can offer concrete contributions to address pressing challenges of TAI in healthcare. These challenges include achieving robustness, reproducibility, and replicability; late-stage development and the integration of AI into clinical practice; the selection of appropriate evaluation metrics; specific AI-related challenges in preclinical and biomedical research; gaps of transparency in medical AI, as well as the need for improved conceptual clarity and AI literacy among stakeholders. Finally, we offer a catalog of ideas and roadmap for future research to inform scholars in both fields on existing interconnections and serve as a foundation for guiding future interdisciplinary efforts.

Ralf Beuthan, Megan Coffee, Heejin Kim et al.

The polygenic risk scores (PRS) have emerged as an important methodology for quantifying genetic predisposition to complex traits and clinical disease. Significant progress has been made in applying PRS to conditions such as obesity, cancer, and type 2 diabetes (T2DM). Studies have demonstrated that PRS can effectively identify individuals at high risk, thereby enabling early screening, personalized treatment, and targeted interventions for diseases with a genetic predisposition. One current limitation of PRS, however, is the lack of interpretability tools. To address this problem for T2DM, researchers at the Graduate School of Data Science at the Seoul National University introduced eXplainable PRS (XPRS). This visualization tool decomposes PRSs into gene-level and single-nucleotide polymorphism (SNP) contribution scores via Shapley Additive Explanations (SHAP), providing granular insights into the specific genetic factors driving an individual's risk profile. We used a co-design approach to assess XPRS trustworthiness by considering legal, medical, ethical, and technical robustness during early design and potential clinical use. For that, we used Z-inspection, an ethically aligned Trustworthy AI co-design methodology, and piloted the Council of Europe's Human Rights, Democracy, and the Rule of Law Impact Assessment for AI Systems (HUDERIA) (Council of Europe (CAI) 2025). The findings of this use-case comprise a comprehensive set of ethical, legal, and technical lessons learned. These insights, identified by a multidisciplinary team of experts (ethics, legal, human rights, computer science, and medical), serve as a framework for designers to navigate future challenges with this and other AI systems. The findings also provide a useful reference for researchers developing explainability frameworks for PRS in diverse clinical contexts.

John Brandt Brodersen, Ilaria Amelia Caggiano, Pedro Kringen et al.

Assessments of trustworthiness have become a cornerstone of responsible AI development. Especially in high-stakes fields like healthcare, aligning technical, evidence-based, and ethical practices with forthcoming legal requirements is increasingly urgent. We argue that developers and deployers of AI systems for the medical domain should be proactive and take steps to progressively ensure that such systems, both those currently in use and those being developed or planned, respect the requirements of the AI Act, which has come into force in August 2024. This is necessary if full and effective compliance is to be ensured when the most relevant provisions of the Act become effective (August 2026). The engagement with the AI Act cannot be viewed as a formalistic exercise. Compliance with the AI Act needs to be carried out through the proactive commitment to the ethical principles of trustworthy AI. These principles provide the background for the Act, which mentions them several times and connects them to the protection of public interest. They can be used to interpret and apply the Act's provisions and to identify good practices, increasing the validity and sustainability of AI systems over time.

Emmanuel Raj, Magnus Westerlund, Leonardo Espinosa-Leal

In recent years we have witnessed a boom in Internet of Things (IoT) device deployments, which has resulted in big data and demand for low-latency communication. This shift in the demand for infrastructure is also enabling real-time decision making using artificial intelligence for IoT applications. Artificial Intelligence of Things (AIoT) is the combination of Artificial Intelligence (AI) technologies and the IoT infrastructure to provide robust and efficient operations and decision making. Edge computing is emerging to enable AIoT applications. Edge computing enables generating insights and making decisions at or near the data source, reducing the amount of data sent to the cloud or a central repository. In this paper, we propose a framework for facilitating machine learning at the edge for AIoT applications, to enable continuous delivery, deployment, and monitoring of machine learning models at the edge (Edge MLOps). The contribution is an architecture that includes services, tools, and methods for delivering fleet analytics at scale. We present a preliminary validation of the framework by performing experiments with IoT devices on a university campus's rooms. For the machine learning experiments, we forecast multivariate time series for predicting air quality in the respective rooms by using the models deployed in respective edge devices. By these experiments, we validate the proposed fleet analytics framework for efficiency and robustness.

John Wickström, Magnus Westerlund, Göran Pulkkis

Proliferation of IoT devices in society demands a renewed focus on securing the use and maintenance of such systems. IoT-based systems will have a great impact on society and therefore such systems must have guaranteed resilience. We introduce cryptographic-based building blocks that strive to ensure that distributed IoT networks remain in a healthy condition throughout their lifecycle. Our presented solution utilizes deterministic and interlinked smart contracts on the Ethereum blockchain to enforce secured management and maintenance for hardened IoT devices. A key issue investigated is the protocol development for securing IoT device deployments and means for communicating securely with devices. By supporting values of openness, automation, and provenance, we can introduce novel means that reduce the threats of surveillance and theft, while also improving operator accountability and trust in IoT technology.

Leonardo A. Espinosa Leal, Magnus Westerlund, Anthony Chapman

Industry has always been in the pursuit of becoming more economically efficient and the current focus has been to reduce human labour using modern technologies. Even with cutting edge technologies, which range from packaging robots to AI for fault detection, there is still some ambiguity on the aims of some new systems, namely, whether they are automated or autonomous. In this paper we indicate the distinctions between automated and autonomous system as well as review the current literature and identify the core challenges for creating learning mechanisms of autonomous agents. We discuss using different types of extended realities, such as digital twins, to train reinforcement learning agents to learn specific tasks through generalization. Once generalization is achieved, we discuss how these can be used to develop self-learning agents. We then introduce self-play scenarios and how they can be used to teach self-learning agents through a supportive environment which focuses on how the agents can adapt to different real-world environments.

Magnus Westerlund, Nane Kratzke

This review focuses on the evolution of cloud computing and distributed ledger technologies (blockchains) over the last decade. Cloud computing relies mainly on a conceptually centralized service provisioning model, while blockchain technologies originate from a peer-to-peer and a completely distributed approach. Still, noteworthy commonalities between both approaches are often overlooked by researchers. Therefore, to the best of the authors knowledge, this paper reviews both domains in parallel for the first time. We conclude that both approaches have advantages and disadvantages. The advantages of centralized service provisioning approaches are often the disadvantages of distributed ledger approaches and vice versa. It is obviously an interesting question whether both approaches could be combined in a way that the advantages can be added while the disadvantages could be avoided. We derive a software stack that could build the foundation unifying the best of these two worlds and that would avoid existing shortcomings like vendor lock-in, some security problems, and inherent platform dependencies.