I don’t get it. The video mentioned was just about text recognition, something AI has mastered long ago. It was not about objects, movements or other complex ac

I don’t get it. The video mentioned was just about text recognition, something AI has mastered long ago. It was not about objects, movements or other complex actions (drawing or building for example). What is so impressive about it then?