The sudden prevalence of Natural Language Processing (NLP) systems, such as ChatGPT, creates a need for resources to support teachers and students in navigating their use in the classroom. While HCI research has contributed hands-on activities supporting AI literacy, NLP is less explored, and there is a lack of research on how to integrate these tools into teachers’ practices. To address this, we conducted an 18-month Participatory Design process with high-school teachers comprising prototyping, workshops, classroom teaching, and dissemination activities. Through this process, we developed and contribute The Engine Room, a public website with novel digital and unplugged NLP activities that are modular and built around customisable datasets. Our findings show how participatory methods can be used to empower teachers in integrating AI literacy activities into their teaching across various subjects and with a sustained impact. We discuss how a participatory approach can address challenges of designing AI literacy tools and activities.

In this paper we argue for broken-hammer pedagogy: a strategy using deliberate breakdowns to reveal core elements of emerging technologies to students. Using AI literacy activities as a case study, we performed an 8-hour intervention in which grade 7 and 8 students (n=34) built, broke, and reflected on simple language models through deliberate and orchestrated technical breakdowns. We coded video recordings of the intervention and conducted subsequent interviews with teachers and students. With these findings, we present three ways in which the technology breaks: Conspicuous, Obtrusive, Obstinate; and reflect upon the learning opportunities and challenges each of these afford. Finally, we demonstrate how students could successfully transfer their learning from the activities to technologies they encountered as part of daily life and discuss how deliberately breaking Heidegger’s famous hammer provides opportunities for exposing students to elements of black-box technology as part of Computational Empowerment education.

LLM-infused tools have entered the newsroom, transforming journalistic work practices. A few studies have investigated how LLMs influence journalistic practices, but there is a lack of research on how to design LLM-infused tools to support trustworthy journalism. In this paper, we explore how prototyping within a defined design space can help identify and explore the challenges of trustworthy LLM integration in journalism. We conduct eight interviews with news industry stakeholders and identify five challenges for trustworthy journalism arising with the introduction of LLMs: factuality, neutrality, autonomy, efficiency, and AI literacy. Based on these challenges, we map a design space and iteratively explore four prototypes of interactive interfaces promoting trustworthy LLM-infused journalism, which are evaluated with news industry stakeholders. We discuss opportunities and conflicts within the design space, how interactive interfaces can be used to concretize guidelines for AI use, and challenges in incorporating Explainable AI into everyday tools for journalists.

Figure 1: To support teachers and students in acting skillfully around AI, we explore how AI technologies and methods can interplay with knowledge and skills in existing school subjects.From left to right: Teachers engaging with AI activities as part of a workshop; students discussing sentences which they are annotating to train a language model; a teacher's lesson plan which integrates machine learning activities into mathematics.