Participants film themselves "speedrunning" through the building, or aiming to complete a task as fast as possible per the common video game slang. The task is to map out Scientology's buildings and get as much information as they can about the inner workings of the organization
I understand the need for processing power, but I thought AI models were just gigs?