The platters have less ability to stabilize platter speed when they are low mass and higher bearing friction. They move quicker because of less inertia, and they are more susceptible to influence from the drive. That is how I understand it. You asked how the AS 2000 maintains speed. The platter and bearing are a part of it. I would not say the same about my former SME turntables or my Denon DD, or the Garrard idlers I have seen. In those cases, the motor and controller have greater influence on final platter speed.
I am happy to be corrected and learn from someone who knows more than I do on this subject. I related to you my direct experience from living with two SMEs, a DD Denon, the Micro, the AS 1000 and the AS 2000 from setting them up, experimenting with thread/belt tension, and observing bearing friction all while measuring results with my Timeline strobe and some with the Roadrunner Tach. If the issue is maintaining smooth, constant speed, the American Sound tables were better than those others, in part because of the platter mass and low bearing friction. I suspect the platter mass and bearing type have other advantages in terms of noise also, but that is a different topic.
To improve on the AS2000 performance, David too his new design to extremes. The platter is wider and taller, for about twice the mass and greater interia. The bearing is similar. The motor is the same but the controller is improved as is the motor housing.
Other typologies also maintain accurate and consistent speed like my Denon and my friend's SP10 MKIII. There are lots of approaches. Good speed is but one of the elements for good sound.
You should interview some turntable designers. Ask them these questions. It would be fascinating and I am sure very educational.