this article is a bit tough... feels like a marketing piece for some Google report.
did I read that wrong or was this whole analysis based on percentages... like what does 76% -> 24% drop in memory related bugs mean in terms of nominal bugs or nominal bugs / kloc
also, it mostly credited memory safe languages but then also just threw this out from the Google report
Based on what we've learned, it's become clear that we do not need to throw away or rewrite all our existing memory-unsafe code
tl;dr android may be producing less memory-related vulnerabilities and it's not exactly clear how