Large Language Models (LLMs) are expected to significantly impact various socio-technical systems, offering transformative possibilities for improved interaction between humans and technology. However, their integration poses complex challenges due to the intricate interplay between societal structures, human behaviour, and technological innovation. This research explores these multifaceted challenges, emphasising the need for a human-centered approach in integrating LLMs to ensure that technological advancements are aligned with ethical standards and societal needs. Utilizing a structured methodology comprising a workshop, literature analysis, and expert collaborations, the study uses a multi-dimensional human-centered AI framework to guide the responsible integration of LLMs. Key insights include the importance of inclusive data, considering unintended consequences, maintaining privacy, and respecting intellectual property rights. The paper identifies and advocates for principles like human-in-the-loop, continuous longitudinal studies, proactive awareness campaigns, and regular audits to develop LLMs that are ethically sound, adaptable, and effectively integrated into various socio-technical systems, thus addressing user needs and broader societal impacts. The paper also underlines the importance of collaboration among academia, industry, and policymakers to develop LLMs that are ethically aligned, socially beneficial, and adaptable to future societal needs. The findings offer valuable insights into the strategic integration of LLMs, advocating for a broader research perspective beyond industrial motivations to fully understand and leverage LLMs in socio-technical landscapes.

As agile manufacturing expands and workforce mobility increases, the importance of efficient knowledge transfer among factory workers grows. Cognitive Assistants (CAs) with Large Language Models (LLMs), like GPT-3.5, can bridge knowledge gaps and improve worker performance in manufacturing settings. This study investigates the opportunities, risks, and user acceptance of LLM-powered CAs in two factory contexts: textile and detergent production. Several opportunities and risks are identified through a literature review, proof-of-concept implementation, and focus group sessions. Factory representatives raise concerns regarding data security, privacy, and the reliability of LLMs in high-stake environments. By following design guidelines regarding persistent memory, real-time data integration, security, privacy, and ethical concerns, LLM-powered CAs can become valuable assets in manufacturing settings and other industries.

Many industries are facing the challenge of how to capture workers' knowledge such that it can be shared, in particular tacit knowledge. The operation of complex systems such as a manufacturing line is knowledge-intensive, especially if the operator must frequently reconfigure it for different products. Considering the breadth and dynamic nature of this knowledge, existing solutions for sharing knowledge (e.g., word-of-mouth, issue reports, document creation, and decision support systems) are inefficient and/or resource-intensive. Conversational user interfaces are an efficient way to convey information that mimics the way humans share knowledge; however, we know little about how to design them specifically for this purpose, especially regarding tacit knowledge. In this work, my main goal is to investigate how a cognitive assistant can be designed to facilitate (tacit) knowledge transfer between users of dynamic complex systems. I aim to achieve this by outlining the design requirements, challenges, and opportunities in factories; by collaboratively designing, implementing, and evaluating a cognitive assistant for sharing knowledge; studying the effects of design characteristics on aspects such as user experience; and finally, creating a set of design guidelines.

Many industries face the challenge of capturing workers' knowledge to share it, particularly tacit knowledge. The operation of complex systems such as a manufacturing line is knowledge-intensive. Considering this knowledge's breadth and dynamic nature, existing knowledge-sharing solutions are inefficient and resource intensive. Conversational user interfaces are an efficient way to convey information that mimics how humans share knowledge; however, we know little about how to design them specifically for knowledge sharing, especially regarding tacit knowledge. In this work, we present an intelligent assistant that we have developed to support the elicitation of tacit knowledge from workers through systematic reflection. The system can interact with workers by voice or text and generate visualizations of shop floor data to support reflective prompts.

Emoji have become an essential part of modern communication, helping to convey emotions and tone quickly and concisely. Emoji used by humans and Intelligent Agents (IA) have been shown to affect people’s decision making intentions, suggesting they could be used to manipulate users to follow their advice. We present a mixed-methods crowdsourcing study (N = 194) that shows that adherence to an IA’s recommendation and user experience are not affected by emoji when used in a positive, collaborative way. However, we demonstrate that explanations provided by an IA do increase adherence to its recommendation.

Learning to operate a complex system, such as an agile production line, can be a daunting task. The high variability in products and frequent reconfigurations make it difficult to keep documentation up-to-date and share new knowledge amongst factory workers. We introduce CLAICA, a Continuously Learning AI Cognitive Assistant that supports workers in the aforementioned scenario. CLAICA learns from (experienced) workers, formalizes new knowledge, stores it in a knowledge base, along with contextual information, and shares it when relevant. We conducted a user study with 83 participants who performed eight knowledge exchange tasks with CLAICA, completed a survey, and provided qualitative feedback. Our results provide a deeper understanding of how prior training, context expertise, and interaction modality affect the user experience of cognitive assistants. We draw on our results to elicit design and evaluation guidelines for cognitive assistants that support knowledge exchange in fast-paced and demanding environments, such as an agile production line.

Conversational agents (CA) occasionally fail to understand the user's intention or respond inappropriately due to natural language complexity. These conversational breakdowns can happen because of low intent and entity prediction confidence scores. A promising repair strategy in such cases is that the CA proposes to users likely alternatives to proceed. If one of these options matches the user's intention, the breakdown is repaired successfully. We propose that successful repairs should be followed by a learning mechanism to minimize future breakdowns. After a successful repair, the CA, user, or both can learn each other's specific phrasing. This prevents similar phrasings from causing reoccurring breakdowns. We compared user preferences for these learning mechanisms in a scenario-based study with manufacturing workers (). Our result showed that users first prefer to share the learning burden with the CA (61.3%), followed by entirely outsourcing the learning burden to the CA (60.7%) as opposed to themselves.

Maintaining a complex system, such as a modern production line, is a knowledge-intensive task. Many firms use maintenance reports as a decision support tool. However, reports are often poor quality and tedious to compile. A Conversational User Interface (CUI) could streamline the reporting process by validating the user's input, eliciting more valuable information, and reducing the time needed. In this paper, we use a Technology Probe to explore the potential of a CUI to create instructional maintenance reports. We conducted a between-groups study (N = 24) in which participants had to replace the inner tube of a bicycle tire. One group documented the procedure using a CUI while replacing the inner tube, whereas the other group compiled a paper report afterward. The CUI was enacted by a researcher according to a set of rules. Our results indicate that using a CUI for maintenance reports saves a significant amount of time, is no more cognitively demanding than writing a report, and results in maintenance reports of higher quality.