The brass is effectively "flung out" with the same speed of the thing that's carrying them when the case mouth
clears the front of the receiver cutout. So ejection speed becomes effectively the same as bolt speed at that point.
And if the bolt is being over-driven relative to restraining forces (i.e, the buffer spring), well.....
I guess you learn something new whenever you are open to learning. I can see that bolt speed would affect ejection velocities if we were talking about a fixed ejector such as a Kalashnikov or a HK 91, but I'll take your word for it with respect to bolt speed and the AR type rifle, even though it doesn't really make sense to me.
I can see where as bolt velocity increases, the spring tension under the AR ejector might become important as there is less "dwell time" where the ejection port is open to allow the empty to be kicked out. But the transition from linear bolt velocity to tangential case velocity with the bolt face ejector button being spring loaded like that...
In essence, if one were to remove the ejector spring altogether, there would be no ejection from the AR. If one were to put a mild spring in there, ejection would be slow, but it would happen, if one were to put a stout spring in there ejection would be more vigorous, regardless of bolt speed.