I will correct you, there have been some incidents one of which I was told was at NASA. But what you and others keep saying which is all anecdotal but yet the CGA standards have not change. So there is more to it than you think.
I don't have a dog in your fight here, but your own reference is also...anecdotal. Do you have a reference to the incidents involving banked 40% EAN, which have concluded beyond a reasonable doubt, based solely on the evidence provided to them, that the cause was the enriched air beyond beyond 23.5% and at or below 40%, premixed?
You keep referencing CGA and how their standards haven't changed, and while it is a valid point, most standards don't change if a tiny subset of users request/demand it, or the same subset of users want to have a discussion about changing the standard. Much like in my own field where I help develop standards, the tiny subset of users won't influence it; however, it doesn't mean there won't be other vendors who create proprietary solutions based on the standard to serve that niche.
The US Navy, the USCG, and I believe even OSHA say 40% is the limit (I believe NASA does too, but I am just not willing to search for their documentation). I've heard, or maybe read, Luxfer requires(ed) cleaning beyond 23.5%, I dive Faber tanks and all operations I've ever used also have Faber HP steel tanks. In fact, remembering NOAA also has skin in this game, I found their guide too:
http://www.omao.noaa.gov/sites/default/files/documents/NDSSM Final_041217.pdf
It seems only ANDI (which uses their trade marked term 'safe air') and CGA (who, once again, likely sees SCUBA as a niche market to the overall demands of the compressed air gas market) adhere to this 23.5% 'standard'. From my own experience, niche end users/use cases don't have enough clout or political power to change or turn standards, even within bodies that serve to set the standards, politics still rule.