Not necessarily, as evidenced by TFA, whose scraper still seems to work even after it visits each individual SERP entry to harvest the URLs.

But Kagi’s SERP API provider is going to have to adapt, I guess.