Academa Uses LLMs to Generate Correctable STEM Lecture Videos as Code

Two PhD students, Sina Atalay and Abdullah Geduk, have launched Academa, a platform that generates long-form STEM lecture videos using large language models. Rather than recording traditional video, lectures are written as code and compiled into video using computer graphics and text-to-speech technology. This approach allows errors to be corrected through code edits, meaning all future viewers automatically receive the updated, accurate version. Each lecture also includes an AI chat assistant trained on the video's full content. The founders acknowledge that LLMs can make mistakes but argue their code-based format enables continuous improvement, unlike static recorded lectures.