I see this as an example of confusion between population-level and individual protection. Widespread use of N95s is great at keeping the hospitals from overcrowding, but if somebody is serious about personally avoiding COVID, 95% percent filtration seems rather low (and there's leakage around the sides too.) If they were truly serious about avoiding COVID, they'd use an elastomeric half-face respirator at the minimum (which has the bonus feature of being more comfortable to wear).
The 3M 9210s I’ve been wearing since 2021 have been independelty tested as having >99.5% filtration efficiency at the relevant particle size, and that’s with physical testing so stuff like ability of the make to seal is also being tested.
Zero COVID infections here.
A properly fitted mask shouldn't leak around the sides. The PFF2 (my country's equivalent of N95) mask I use doesn't leak anywhere when properly adjusted. (Like the sibling comment, it's a 3M, either the 9320+BR or the 9360H; see https://www.3m.com.br/3M/pt_BR/p/d/v000465595/ for its page complete with usage instructions.)
If you stick a mask on a dummy in a lab and fire COVID at it, it might show pretty good protection. However, at the population level there is no evidence that masking confers any benefit in preventing the spread of COVID.
The Cochrane Review demonstrated this: https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD...
There are also numerous dis-benefits of masking, especially among young children.
I don't see the point of your comment. You're presenting an anecdotal case where someone who took some precautions which are known for not being 100% effective against airborne diseases ended up contracting an airborne disease.
It makes as much sense as ranting about wearing seatbelts because you can put together an anecdotal case of someone who got hurt in a car accident in spite of wearing one.
What's the point of that sort of argument, really?
The actual precautions necessary to protect him from any exposure at all would have been significantly more stringent than what people tend to believe are reasonable, as GPs anecdote demonstrates.