We are all in big trouble!
Summary
"We become what we express."
Climate change is a source of anxiety about the future. Understanding how people express themselves about climate change enables us to address such concerns. To study climate change expression on social media, we analyzed 200 TikTok videos tagged with #climatechange, identifying four categories of content: expressionfeelings, views-appeals, news-information, and trend-hijacking. We found that creators use humor to package sharp critiques, avoiding direct confrontation. They replace complex discussions with life stories, such as adopting a vegetarian lifestyle or deleting emails. They borrow from news media to present fragmented information as scientific interpretations, creating a perception of scientific credibility, balancing scientific accuracy with emotionality. Analysis of viewer responses showed they engaged empathetically, reshaping interpretations of videos. These interactions risk reinforcing existing views but help build community on TikTok, which lacks community structure. This study reveals how creators may retell news on science using personal narratives, highlighting how short-form videos enable climate communication.
Inspired by this research, we created an artwork called CLICKLOC with social-media sourced videos and AI-visions of what are said. To probe how we express our contemporary anxieties, emotions, storytelling, and interpretations of climate change, we crowd-sourced TikTok videos expressing our relationships to climate change and use GenAI to fill out the stories in visual form. This process reveals how GenAI becomes an expressive agent to express our own often isolated stories and arguments about climate change, but it also cautions us to realize how AI can produce inconsistent and nonrepresentative depictions and can subtly alter our intended forms of expression. The video was made using the Stable Diffusion Webui and the Deforum plugin. The AI-generated visuals on the right side of the video are derived from dialogues in the original left-side tiktok video. We utilized DeepSeek to algorithmically summarize the conversations, then fed the extracted keywords into WebUI to generate corresponding scenes. By juxtaposing the original and AI-rendered videos side by side, this parallel editing technique amplifies narrative coherence and intensifies the atmospheric impact of the visuals. The video's opening and closing segments mimic the format of TikTok clips to reinforce a casual, accessible vibe. We first segment each sentence of Tiktok videos, and use Deepseek to extract words that may be converted into visual elements in each sentence, and convert the words into prompt words with a sense of the storyboard, then input the prompt words into Deforum to generate AI videos, ensuring that the generated video has the same rhythm as the original Tiktok video.
Publication: Proceedings of the ACM on Human-Computer Interaction (CSCW'26), arxiv.
Exhibition: website.
Shown at HABA Art Lab, Barcelona, Spain, 22 April to 1 May 2026; Arts Itoya, Takeo, Kyushu Japan, 22 June 2025; TESA Taiwan Environmental Sculpture Association Creative Center, Taiwan, 10 Jan 2026; School of Creative Media, City University of Hong Kong, 15 Dec 2025.