> Accepting the built-in device software is also an "experiment."

An experiment with substantial testing and one that is legally backed if something is wrong. These are class 3 FDA approved devices, which means they have rigorous testing. They aren't just FDA cleared.

So calling them an "experiment" is really understating it. There's a huge difference between an experiment with one participant and done by someone without formal medical knowledge and many experiments with large sample sizes performed by several medical professionals who are experts on the specific device at hand. The two aren't comparable, and it's a ridiculous notion to equate them.

In the context of my statement, being conservative means accepting lower risk.

You're looking at the risk of the software failing, and I'm talking about the risk of the entire system. Which includes the kid who has to control the pump, and we are not the experts on the kid.