I thought it is by introducing an RCE vulnerability that you get an RCE vulnerability.

I'm being facetious of course, but this recent rhetorical trend of people confidently vouching for "pet" in "pet vs. cattle" is not a sustainable decision, even if it's admittedly plain practical on the short to medium run, or in given contexts even longer. It's just a dangerous and irresponsible lesson to blindly repeat I think.

Change happens. Evidently, while we can mechanistically rule out several classes of bugs now, RCEs are not one of those. Whatever additional guardrails they had in place, they failed to catch this *. I think it's significantly more honest to place the blame there if anywhere. If they can introduce an RCE to Notepad *, you can be confident they're introducing RCEs left and right to other components too **. With some additional contextual weighting of course.

* Small note on this specific CVE though: to the extent I looked into it [0], I'm not sure I find it reasonable to classify it as an RCE. It was a UX hiccup, the software was working as intended, the intention was just... maybe not quite wise enough.

** Under the interpretation that this was an RCE, which I question.

[0] https://www.zerodayinitiative.com/blog/2026/2/19/cve-2026-20...