A few incidents with excessively dangerous stuff like smallpox aside, most laboratory security incidents are essentially no worse than what millions of people experience on any given day. Scratched by sick animals, tick bites, mosquitos, bites larger animals, eating sick animals, getting blood from a different animal in an open (perhaps tiny) wound, breathing the air of densely held animals, leeches, etc etc etc.

I mean, not saying labs shouldn't be extremely careful, especially if the pathogen isn't unusually dangerous - but the outside world generally contains quite a few dangerous pathogens too, any more more (and more intense) interactions between host and human in which bad things can happen.

To be clear, these issues are serious and need addressing, but nevertheless the sheer numbers involved would suggest that the security incidents by themselves are unlikely to cause this kind of pandemic. We'd need some additional ingredient, like intention infection with multiple human subjects over a prolonged period, or extremely successful gain-of-function... something that can shift those otherwise billion-to-one odds very, very far.

Out of a kind of morbid fascination if indeed it turns out this long shot actually happened it would be really interesting to read the post-mortem as to which factors contributed, because it kind of has to be a list of bad luck and poor decisions.

(But to reiterate: it's not OK to ignore risks even if they aren't the most likely source of pandemic outbreaks, that's a way too low bar!)