at least this way you’re getting a signal - “Location is not to be relied on”
I often have the opposite problem. People giving a location accuracy of 10m, and when I chase them up, only then, I find the animal was 7km or 15km away. So I’d prefer a huge ‘accuracy’ circle.
Assuming 15m accuracy from any (non-professional equipment-based) method for determining coordinates while hiking through diverse habitats/geography seems likely to overestimate precision (i.e., err on the side of too high precision) in at least some cases. Both GPS and phone-based (especially phone-based) location errors are reasonably frequently above 15m in my experience, and I’m not generally in particularly demanding conditions (like canyons/ steep valleys, etc.).
That’s fair, I could well go higher. Generally I do a quick check to make sure it makes sense, and fix individual observations that don’t. If the camera time is off by a couple minutes it can make a difference as well especially if I’m moving, although at iNatting pace it’s usually not a big difference.
Unlike “1.6525m” or “150m”, picking a default ‘accuracy’ value of “15m” (whatever statistic is used) raises no alarm for me; the 15–25m range is realistic and on the safe side, if using a fairly recent consumer-grade gps device in normal conditions. (N.B. People passionate about geodesy or after estimating the truth-to-ground of their equipment will find various guides and apps on the internets)
I’d be waaaaaay more anxious about a 2-mn desynchronization between gps and camera 
(Unless some record is kept, in order to compensate later for the nasty timeshifts.)