It's certainly something which gets studied, for example, this meta-analysis of other studies in 2014 suggests a J shaped curve in T2s, and it seems that what they mean by that is very high a1c poses a very high risk, and very low a1c poses a risk but not as high a risk as high a1c.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4310064/
I've no particular view on it either way; a1c is getting kinda irrelevant for cgm users: I think our aim is to spend each day within a decent range with modest standard deviation and let the a1c fall where it may.
Gut instinct does make me feel, though, that there are diminishing returns and increased risks in chasing an ever lower a1c just for the sake of it - there's bound to be an "ok-ish" a1c range and getting any lower seems not to bring any added physiological benefits.
I have no view on what that range is.