Developing research indicates that teenagers have been integrating generative AI chatbots into their social lives through various ways—from enlivening group chats to nurturing intimate companionships. However, we know little about how these AI chatbots may shape the interactions between teenagers themselves and their close friends, despite the important consequences adolescent peer friendships have on social functioning in adulthood. To develop our understanding of this critical area, we conducted a speculative enactment and interview study with 47 teenagers aged 13-19 to understand teenagers’ views of AI chatbot use in the context of navigating consequential friendship interactions. After enacting speculative scenarios involving their own close friend, participants’ reflections highlighted distinct perceptions about the meaning of friendship, human effort, and interpretations of acceptable AI use. We discuss considerations for the impact of AI chatbots on teens’ social development and present design implications to support teenagers’ relationship autonomy and active social learning.

Learning is essential for the long-term sustainability of apprenticeship-based communities; it lets novices gradually progress to experts through active engagement. However, the use of large language models (LLMs) can disrupt this trajectory: if novices rely on LLMs to generate content, they risk short-circuiting the learning opportunities embedded in the process of creating content themselves. To address this tension, we created WikiCoach, a Q&A–style LLM system designed to scaffold learning-through-doing in Wikipedia. WikiCoach does not generate content directly; instead it supports editors via three stages: 1) introducing relevant community policies and guidelines, 2) operationalizing them into actionable steps, and 3) providing evaluative feedback on initial drafts. We conducted a within-subjects study (N=28) comparing WikiCoach with a baseline content generation model, and found that WikiCoach significantly improved policy understanding and adherence to community norms, while increasing critical thinking and cognitive engagement. Participants also preferred WikiCoach for future editing tasks. Qualitative analysis revealed how WikiCoach shaped users’ expectations and interactions. Our findings demonstrate that AI systems can simultaneously support task completion and learning, offering a future direction for responsibly sustaining apprenticeship-based communities in the era of generative AI.

In China, stay-at-home moms (SAHMs) often experience a hard time during their transitions, such as returning to school or restarting their careers. Yet, their experiences throughout this transition and their strategies to overcome potential challenges are rarely studied in the CSCW literature. In this study, we examined how Chinese SAHMs leveraged social media platforms to assist their transition and help them navigate the potential challenges. Through interviewing 15 SAHMs who have successfully completed their transitions, we identified the key role that social media platforms (e.g., RedNote, Douyin) play in helping SAHMS build resilience as they re-engage with society. For example, many SAHMs found peer support during their transition from mom-centric social groups where they could receive advice that was specific to their situations and constraints. When they achieved their goals, they turned into contributors in such groups and altruistically shared their experiences to help other moms. Based on the findings, we discussed the opportunities to support SAHMs' transitions.

Intangible Cultural Heritage (ICH) faces numerous threats that can lead to its destruction. While the emergence of short video platforms provides opportunities for fostering innovation and communication among ICH practitioners and viewers, it is still understudied how different stakeholders present, explain, and manage ICH via short videos. To address this, we conduct a mixed-method study of ICH-related videos on Douyin, a popular short video platform in China with an extensive user base and wealth of ICH content. By adopting the Critical Heritage Studies (CHS) framework, we propose a taxonomy of frames that construct the landscape of ICH short videos and then investigate the interactions among different groups regarding power, identity, and knowledge. Additionally, we analyze viewer responses to different frames and groups based on audience metrics (e.g., # of likes and comments) and comments. Our research reveals that government-affiliated and indigenous groups dominate the promotion and presentation of ICH on Douyin. Contrary to previous literature, viewer responses show a preference for videos from external ICH groups and ordinary individuals, suggesting a tendency to counter authority and exclusivity associated with ICH. Moreover, it highlights a lack of sustainable debates and negotiations among different groups involved in ICH discourse. Situated within CHS, we provide design implications for ICH safeguarding and sustainability through short videos and online media.

Sentiment analysis is an important and challenging task in natural language processing. It has been studied for a few decades. Recently, Bidirectional Encoder Representations from Transformer (BERT) model has been introduced to tackle this task and gain very promising results. However, most existing studies on fine-tuning BERT models for sentiment analysis focus on high-resource language (e.g., En-glish or Mandarin). This paper studies the sentiment analysis of Cantonese political posts on Hong Kong local forums. We first collected and labeled posts related to Anti-Extradition Law Amendment Bill (Anti-ELAB) movement in Hong Kong discussion forums. We then examined the performance of dictionary-based sentiment analysis, traditional machine learning-based, fine-tuned BERT and fine-tuned multilingual BERT (mBERT) models. Our results show that fine-tuned mBERT model achieves the best performance on our collected and labeled Cantonese dataset.

Preventing the spread of fake news is one of the most challenging issues in the age of social media. Traditional manual fact-checking (i.e., expert-based and crowd-sourced fact-checking) is time-consuming and labor-extensive, which cannot scale up with the unprecedented amount of dis- and mis-information on social media. Automated fact-checking based on machine learning is a promising strategy to address the scalability issues. Nevertheless, an end-to-end full automated fact-checking system without human supervision is still impractical. A more realistic solution will be developing an Artificial Intelligence (AI)-based system to facilitate the human fact-checkers during the fact-checking process. Therefore, this paper proposes a novel annotation system to facilitate human fact-checkers. With our designed procedures and schema, our developed system can help to improve the efficiency and effectiveness of human fact-checkers by automatically identifying worth-to-check news. We conduct a real-case study to demonstrate that our system can effectively identify worth-to-check news and ease the annotation process with the help of several automatic detection functions.

With the prevalence of social media, fake news has become one of the greatest challenges in journalism, which has weakened public trust in news outlets and authorities. During the COVID-19 epidemic, the widely circulated pandemic-related fake news on social media misleads or threatens the public. Recent works have investigated fake news detection on social platforms in English and Mandarin, though Cantonese fake news has been understudied. To pave the way for Cantonese COVID-19 fake news detection, we first presented an annotated COVID-19 related Cantonese fake news dataset collected from a popular local discussion forum in Hong Kong. Then, we explored the dataset by applying topic modeling to identify the topics that contain the most significant amount of fake news. Moreover, we evaluated both traditional machine learning algorithms and deep learning algorithms for Cantonese fake news detection. Our empirical results show that deep learning based methods perform slightly better than traditional machine learning methods on TF-IDF features.