Survivorship aka selection bias only exists if you are drawing biased or erroneous conclusions. Defined as "deriving an erroneous assumption because dead people aren't reporting"
If 100% of people dive a 6lbs axial for 4 hrs in local waters (45 to 55F) and there are zero Co2 hits (going to ignore packing errors or time counting errors) there is no error there. So you are not using the reporting "survivors" to make an erroneous conclusion. Is there something to overlook is an open question.
Survivorship bias - Wikipedia
"
Survivorship bias or
survival bias is the
logical error of concentrating on the people or things that made it past some selection process and
overlooking those that did not, typically because of their lack of visibility."
To confirm we had/have selection bias we have to look for and find a dive (preferably more than one) with a co2 hit, in under 4 hrs. There are two problems with this. The first is that its way outside the norm aka rare like way more than 3 sigma rare. So "has it happened" is unknown, "was it reported" and who it was reported to also unknown. There's no master database of dive vs Co2 events so proving that negative isn't really possible. Either that lone Co2 hit doesn't exist, it isn't reported, it was reported and hushed up, maybe it was reported just not to "us" or publicly - are all possibilities that we can't conclusively know. But not having data is not the same as overlooking contrary data through erroneous selections.
The other issue is what statistical risk are you willing to accept. 4 sigma below the curve would be ~6 CO2 hits in 1,000 dives which seems like we would have heard about since the Meg axial has been in production since about 1996. But that assumes that scrubber breakthrough at the extreme tails of its time distribution is normal - which it probably isn't even under one set of test conditions nevermind under a myriad of depth, time, co2 load, and diluent conditions. Going to guess here that at some point the tails of the distribution have become such rare conditions that there aren't enough CCR dives in recorded history to have a Co2 hit that soon or to have a scrubber last that long (two tails). The CE test data don't really inform us about those tails, so we are left making inferences.