When a voiced consonant (b,d,j) is followed by an unvoiced consonant (s,f,t) we’re used to pronounce it as un unvoiced one (“p” instead of “b”, “sh” instead of “j”) but as far as I know, it’s not a rule, it’s rather a phonetic phenomenon.
Actually, it’s not impossible to pronounce a clear “b” in “absent”, “abscons”, “obscurité”; but it would be like passing on a speed bump. Pronouncing a “p”, in this case, is easier, and it becomes a habit, even more so when you’re fluent in french.

I personally pronounce these as /bs/ not /ps/, but I’ve heard it both ways and my sense is that most native speakers wouldn’t be able to tell the difference if they heard the same word pronounced both ways. As to the original question of whether this is only at the beginning of words — no, there’s no difference in how these clusters are pronounced between initial and non-initial placement such as “nonobstant”, “réabsorption”, etc.