Software is inherently complex and as a result over the years we have spent significant resources designing techniques for automated testing, debugging and repair to help ensure its correctness. Some of these techniques leverage algorithms that mimic biology, a natural domain with built in complexity, from which our community has made many parallels. These testing techniques are often predicated on the fact that we have the ground truth and a single set of specifications, and that the system behaves deterministically. However, the software development process and types of software we are building today is rapidly changing and these assumptions may no longer hold. In fact, our software is becoming more organic, resembling the biology we sometimes exploit to test it. In this talk I discuss some forays into software testing in emerging and scientific domains where the boundaries of our assumptions are becoming fuzzy and discuss a future of software testing within this context.