I think you nailed it right there.
What we're seeing is a difference between professional and lay reporting, similar to how medical journals and blogs reporting on their articles sometimes differ (Nature's "Myocarditis among Covid-vaccinated patients" being among the most famous examples of what can go wrong when lay people pass around literature written for specialists).
Although computer scientists will use a term that sounds like an objective Boolean, among their peers it's understood to be a subjective range.
Twitter is already aflame with arguments from scripters with no formal CS training over this, and I'd guess Reddit is no better.
That we should see a perceived conflict where there is none in this forum is unsurprising.
Scripting has so many productivity benefits over the tedious bit-counting of lower-level languages, it hardly needs memory management security to make its case.
Long before this report, Python had already become the most widely used programming language, at least per the TIOBE Index.
In fact, my question would be based on productivity metrics alone:
Each language has a place. You probably wouldn't want to write an OS kernel or device driver in Python. But in the cases where Python is a good fit but the team is using C, why burn the extra payroll?
.