As the article is a bit short on details we can only grasp at what they exactly do different - maybe they're not doing anything unique, maybe he's just impresse

As the article is a bit short on details we can only grasp at what they exactly do different - maybe they're not doing anything unique, maybe he's just impressed with how they're using "known" methods to minimize risks/randomness. I really can't tell from the text what kind of cells they use and how random these are, or how novel their method is (if it is even novel).