State-of-the-art intelligent and interactive agents, such as Alexa or Siri, often present overly conforming behaviour during interactions with humans. This can result in a misalignment between end-user expectations and agent behaviour. To overcome this barrier in human-AI interactions, we introduce the Critical Friend (CF), a conceptual idea that guides critical behaviour in human-human interactions. We present our results as a formal understanding that can be described through description logic and utilised for reasoning capabilities, enabling implementations of the CF as an intelligent interactive agent.
Expanding electrification infrastructure demands route planning tools that optimize feasibility while respecting legal, environmental, and stakeholder constraints. We present a hybrid system for value-centric electrification planning that combines heuristic search with logic-based verification. A modified A* algorithm explores alternatives using expert-informed heuristics, while an Answer Set Programming (ASP)-based rule set evaluates stakeholder constraint compliance. This separation of planning and evaluation supports transparency and explainability. Expert feedback highlights the system’s value in early-stage decision support, and statistical analysis suggests that it consistently identifies technically feasible routes with reduced private land intrusion.
This paper presents an ontology-based approach to proactive risk assessment in high-risk industrial environments, with a focus on molten metal handling in smelting plants. Distributed cognition theory serves as the theoretical framework, guiding the grounded theory and thematic analysis by framing collaboration and communication as emerging from interactions among operators, equipment, and the environment. This lens emphasized shared knowledge and distributed responsibilities, helping identify domain-relevant concepts critical for risk assessment. Expert insights from professionals at a Swedish smelting plant were systematically elicited and structured into an ontology to support proactive risk management. A prototype system integrating this ontology with a dynamic interface showed strong alignment with expert decision-making and safety assessments. Moreover, the system identified overlooked risks—such as hazardous equipment containing molten metal—and received positive user feedback. While tailored for smelting operations, the methodology has broader applicability to improving information sharing, decision-making, and safety in other socio-technical systems.