The idea makes sense in some abstract theoretical way, but in practice this will simply reflect the biases of users by representing the subjective "sketchiness" of (typically) poorer areas, leading to gentrified travel routes that siphon yet more business away from communities who could really use it.
My area is probably not representative of most Nextdoor communities however
I am only speaking out because I get the sense from this community (and having worked in the tech industry for long enough) that such externalities are not really considered when designing apps like this one. The other comments give a hint to the proportion of time that these concerns and issues are discussed.
I see all the time only the consideration of "could" without any regard for "should".
I could share with you many stories from FAANG privacy compliance groups about how little engineers think about the impacts of their releases.
while it's important not to conflate "poor" with "unsafe", I don't think it's unreasonable or unfair for individuals to try to avoid walking around in areas with a higher rate of violent crime. as long as the service makes a good faith attempt at objectivity, I don't feel there is anything wrong with a service providing this sort of information. crime maps might be a good place to start looking for data.