Just yesterday I had posted a similar thread where an AV was MITMing traffic between the browser https://news.ycombinator.com/item?id=13307096. I'm still curious why this practice is allowed and these CAs are still trusted by browsers and such.
Many things that would take advantage of those would need to be written to disk and the AV could presumably catch them then, but how about JavaScript exploits? And how does the browser respond to writing things to disk (presumably having exclusive write access) then losing access to them?
There's a juggling act going on there, and I suspect it's a lot easier for an AV vendor to capture and review the network traffic as a single point of contact rather than trying to work with multiple browser and other software vendors to make sure that their software interoperates correctly with the AV.