Again, that happens more often than you think. A key point is whether the original authors can repeat their own work at will. If so, then that is an indication that the second "reproducing group" has one or more variables that are uncontrolled. And those may be variables that the first group is not even aware of, so it may or may not be due to "poorly designed experiments". But the RIGHT way to address the issue is not for the second group to pop up and say "it's a scam", but to work with the first group to see if that/those variables can be identified and controlled.
This is a major reason that various raw materials have "Lot Numbers" associated with particular batches.
"...but I am unaware of a commercial product without any theory going to market. That would seem to be dangerous when the product could be found to either not function as expected or is hazardous."
How about the steam engine?? And it "was" dangerous, as witness the many explosions along the way. Likewise gunpowder. Ditto on the explosions. Firearms?? When you start thinking it through, you realize that much of the technology that humans have today was developed "pre-science" by nothing more than "trail and error".
Gun Powder and Steam Engines all had well proven theories.
I was expecting you to say something like the Dialysis machine when had it’s theory thrown out for a new one years after it was a commercial device.
To my knowledge there hasn’t been such a product that didn’t even have a theory.
If the experiment is only succeeds when run by the original authors by definition it isn’t reproducible.
Are we to say that this device can be commercialized and run by many different individuals when you can’t even get different labs to reproduce each others work? If you can’t even describe how to run the experiment to another scientist, you can’t spec it out so that a manufacture can mass produce it.
This just doesn’t pass the smell test.