[$] Identifying buggy patches with machine learning

The stable kernel releases are meant to contain as many important fixes aspossible; to that end, the stable maintainers have been making use of a machine-learning system to identify patches that should be considered for astable update. This exercise has had some success but, at the 2019 OpenSource Summit Europe, Sasha Levin asked whether this process could beimproved further. Might it be possible for a machine-learning system toidentify patches that create bugs and intercept them, so that thefixes never become necessary?