I don't think 23 is wildly off from 25, and honestly this is just the first one I found that mentions it, I've seen various different sources for different reasons in the past. But the average is based on genetic mutations, and obviously in any given human it's irrelevant how large a generation is as to how much genetic mutation is contributed by the generation. Like even if there are 8 billion people today, that doesn't imply that you somehow got more generic inheritance from your parents than they did from theirs back when there were 6 billion people or whatever. Judging average to be the average per generation (a reasonable inference given the methodology) the last few years won't make much of a difference in a timescale of 250k years

I can't find the article I vaguely remember from a while ago, here's another random one that has mothers in the bronze age ranging from 16-40ish https://www.researchgate.net/publication/314262257_Bronze_Age_Beginnings_The_Conceptualization_of_Motherhood_in_Prehistoric_Europe although you can't really infer much about averages from that.

Anyway yeah there have been periods in time when average age of mothers was younger, but generally if you look back on a long timescale it's been older than people seem to assume. Seems to be quite common to have the notion that women all had children at 16 or whatever back in the day but not much to really bear that out that I can find.