A new AI tool is turning dense, time-consuming academic research papers into short, 45-second TikTok science videos for a general audience.
Here’s the story behind PaperTok. Students at the University of Washington’s (UW) Prosocial Computing Group began to notice a trend on social media: Members of the general public were using generative artificial intelligence to make short science videos. The only problem was that they weren’t scientists—increasing the already high likelihood for mistakes, given it was AI-generated content.
Intrigued, researchers at UW created PaperTok, a platform which uploads a scientific paper and uses Gemini to write a short script, transforming it into a video clip. (Currently, PaperTok is only accessible with a paid Google Gemini subscription.)
“For several reasons, most people don’t read research papers,” senior author Gary Hsieh, a UW professor in human centered design and engineering, said in a release. “I still have challenges reading papers in fields I’m not familiar with. So we wanted to find a way to quickly turn papers into a format that laypeople would want to engage with, and we wanted to study how they engaged with it.”
In April, the team presented their research at the Association for Computing Machinery Conference on Human Factors in Computing Systems in Barcelona.
Here’s how to generate a video using the tool:
1. Upload a PDF and select a script
First, go to the PaperTok site and upload an academic paper. The system automatically analyzes the content and generates four possible options for the video’s hook (or lead). Choose your favorite one. PaperTok then generates a script, which users can edit. Refine the copy and voiceover tone in the editor.
2. Generate video scenes in a storyboard
Once the script is complete, users can generate corresponding video clips, which are broken into scenes, like a movie storyboard. Users can refine and edit the scripts and matching video clips.
3. Add credits and generate the final video
Finalizing the video: The tool automatically generates screen credits from the paper’s original authors, to which you can add your own byline. Last step: Merge all scenes into a shareable short-form video.