Explainable and Trustworthy Intelligent Systems: A Systems Perspective

Dear Colleagues,

The rapid integration of artificial intelligence into critical domains such as healthcare, finance, energy, transport, and public administration has transformed AI from an experimental technology into a decision-making infrastructure that directly affects human lives. This shift has brought explainability and trustworthiness to the forefront of both research and policy agendas. Regulatory frameworks such as the EU AI Act, alongside growing public scrutiny of algorithmic decision-making, are compelling researchers and practitioners to move beyond narrow, model-centric notions of interpretability towards a broader understanding of trust as a property of the whole intelligent system, encompassing data provenance, learning processes, deployment environments, human oversight, and security safeguards.

This Special Issue takes explainability and trustworthiness as system-level properties rather than attributes of an isolated model. A single interpretable algorithm cannot, on its own, guarantee that an AI-enabled service is trustworthy in operation. Trust emerges from the interaction of multiple components across the system lifecycle: the provenance and quality of training data, the learning process and its assumptions, the infrastructure and environment in which the system is deployed, the mechanisms available for human oversight and intervention, and the safeguards that protect the system against adversarial or malicious interference. Treating these elements as an integrated whole, rather than addressing them in isolation, is what distinguishes a systems perspective from conventional explainable AI research, which has tended to concentrate on rendering individual model outputs interpretable. This distinction is central to the scope of this Special Issue and to its fit within the aims of Systems.

At the same time, the increasing complexity of modern AI architectures, including deep learning, federated and distributed learning, and large-scale generative models, has made transparency more difficult to achieve precisely when it is most needed. Challenges such as data privacy constraints, adversarial vulnerabilities, model opacity, and the tension between performance and interpretability remain unresolved. These challenges also present opportunities: advances in explainable AI, privacy-preserving learning, synthetic data generation, and human-centred design offer promising pathways towards systems that are both high-performing and accountable. This Special Issue is therefore timely, offering a platform to consolidate emerging research and to bridge the gap between theoretical advances and the practical realities of deploying trustworthy AI at scale.

This Special Issue aims to present and disseminate the most recent advances in the design, evaluation, and deployment of explainable and trustworthy intelligent systems, viewed from an integrated systems perspective. Rather than treating explainability, security, privacy, fairness, and robustness as separate research strands, we invite contributions that examine how these properties interact, reinforce, or constrain one another across the full AI system lifecycle, from data collection and model development through to deployment, monitoring, and decommissioning. Submissions are encouraged to address the technical, methodological, and socio-technical dimensions of trustworthy AI as they arise from this systems-level interplay.

Dr. Oras Baker
Dr. Zahra Ziran
Dr. Ala Arman
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the Special Issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-anonymized peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Systems is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.