Video is the most consumed medium in the world, yet producing it remains challenging, demanding deep expertise and tedious manual work. Advances in generative AI are rapidly reshaping this process, opening opportunities to compress timelines, blur boundaries between production stages, and enable new forms of storytelling. In this talk, I will present a series of AI systems that support creators across different aspects of video making—from scriptwriting and ideation to editing and feedback. These systems help creators iteratively develop scripts through audiovisual feedback, explore narrative possibilities through generated video, collaborate with AI agents during editing, and reflect on audience responses before publication. Together, these projects illustrate a human-centered approach to AI-assisted video creation that keeps creators in control while providing meaningful automation. I will close by reflecting on the new storytelling affordances generative AI unlocks, alongside the existential questions it raises for the craft itself.
| 11:45am - 12:15pm: | Food and community socializing. |
| 12:15pm - 1:15pm: | Presentation with Q&A. Available hybrid via Zoom. |
| 1:30pm - 2:15pm: | Student meeting with speaker, held in the same location. |
Bryan Wang is a Research Scientist at Adobe Research. His work advances how AI can amplify human creativity and storytelling, with a focus on video creation. At Adobe, he leads projects from exploratory research to shipped products, with work featured in outlets such as The Verge, Fast Company, and Engadget. His research has been published at HCI and AI venues such as CHI, UIST, IUI, AAAI, and EMNLP, and he co-organized the inaugural ICML Workshop on AI & HCI. He received his PhD and MSc in Computer Science from the University of Toronto, advised by Tovi Grossman, and his BSc from National Taiwan University. He has also held research positions at Google Research and Meta Reality Labs Research.