In practice the point of most security research is: get the paper published and give the talk, or sell the zero-days, or use them yourself, and then move on to

In practice the point of most security research is: get the paper published and give the talk, or sell the zero-days, or use them yourself, and then move on to the next thing. The actual non-sexy defensive work of cleaning up bugs and implementing countermeasures, i.e. the real work on improving security instead of making it worse, is mostly done by rank-and-file software and hardware developers whose names are never known.