I don't see how anything in this exploit requires ads. It could just as well be done by enticing the target to view a malicious web page, such as a post in a subreddit or facebook group the target is known to read, or even by just sending him a link in an email or chat which will often trigger an automatic "preview" of the link in the email or chat client. Or by other social engineering techniques.
The ad-tech network probably does make it easier, because any page that includes their ad network could deliver the malicious ad.
The main thing ads allows is "zero-click" - so they don't need to click a link or anything to get exploited.