Of course there will be exceptions, and it wouldn't need to be perfect. It just needs to be better than counting hyphens alone, which the suggestion above inarguably is.

Not to mention fantasy/SF love of invented languages. I'd hate to be HP Lovecraft if this becomes common.

Or have to quote transliterated Arabic, for that matter.

Names are proper nouns and therefore capitalized, the edge case is resolved until you find a proper noun with capitalization in the middle of the word with the hyphen coming before the capital letter. This seems to be limited to foreign words(German in particular) and Company Names where the author inserted a hyphen. https://en.wikipedia.org/wiki/CamelCase#Current_usage_in_nat...

Well then they shouldn't do it at all or rethink their algorithm altogether.

Or do the sane thing most other people do: back up an imperfect but effective algorithm with competent human review.