There are a few problems with this.
First, the punishment varies between students - some will have no problem going in front of the class and doing this (some will view this as an invitation to cheat until caught the first time), others might not be able to stomach the thought of doing that. In my experience working with undergrads, those who feel like they had to cheat, the ones you want to let off with a warning, are much more likely to fall into the second group.
Second, no reputable university would ever allow such a thing. They universally have systems in place dictating exactly what happens when a student is caught cheating. Individual professors don't have leeway in deciding how to deal with these things.
Third, the "make the students police each other" approach never works. The only thing it does well is turn the entirity of the student group against you.
Fourth, I think you make too big a deal out of the repercussions of cheating. You have to bear in mind that the overwhelming majorty, by a landslide, will never be researchers. In fact, depending on your field, it's relatively likely that they'll never do anything even related to their field of study. Those who are going to be researchers might be covering up a particular deficiency. I've known people who were absolute terrors in their field, but just couldn't do arithmetic quickly or couldn't remember formulas without some sort of reference. These are not problems you face in the real world of research. I met one girl who was a very strong philosopher, but had a particular, very rare condition that made her unable to interpret symbolic logic. And then you have the serious people who are forced to take classes that have nothing to do with their field - classes that they might struggle in or might be taking too much time from their actual focus with useless busy-work. Regarding your father's criticism, it's also possible that his expectations are just distinct from modern teaching in the field. This happens very often in technical fields.
And even if they do become researchers, this is why peer review exists, why laboratory reviews exist, et cetera. There are already controls in place to prevent this sort of thing without trying to predict who will end up falsifying data before the fact.
Finally, if they want to become researchers, they have to go to grad school. This is a much better filter than looking for cheating in undergrad careers. It is unbelievably hard to cheat your way through any remotely reputable grad program. You're talking about very personal interactions and production of original research. If you learn enough to "fake" your way through grad school, congratulations, you accidentally became an actual grad student.