Mo Bitar tests a consumer AI assistant after discussing the broader debate over alignment and the pace of development. He asks it to create a shared vote for choosing a game, then examines the difference between generating a form and delivering a working shared service.
Mo Bitar tries multiple conversations, a long planning document, a generated podcast and a satirical video. The fictional podcast voices are generated artifacts, not additional real participants. These examples illustrate his criticism of sprawling outputs and uneven usefulness, not a standardized benchmark.
Mo Bitar encounters login and session problems and questions what would make a general assistant valuable to a large consumer audience. His conclusions are humorous personal commentary based on the demonstrated session. Promotion of his own application is omitted.
Watch on YouTube
