Depends on what you really want to do. If you want to be a life sciences researcher then you need all the expensive equipment that only institutions can afford.
Agreed. The same goes for most engineering fields too.
Software development is one of the few high-tech fields where one can reasonably attempt to get an education through online courses, but it's becoming so commoditized that the bottom is going to drop out and drop out hard, eventually.
Try going to the lab instructor and asking him or her if you can come back later at night to run some of your own experiments. You'll get a hearty laugh.
The expensive equipment becomes an argument in favor of a traditional education only after you get to the post-grad level when you have real access to that equipment.
The second problem with internships instead of school labs is the classic birth problem where we produce too many trained chemists for too few jobs. And flooding the labor market with free intern labor isn't going to help the fraction of grads who can get a job in the field. The result would be even more people with chemistry degrees doing anything other than chemistry as a profession after graduation.
The third problem with internships instead of school is related to the second in that pre-med uses ochem as a weedout, or used to. So "most people" trying to learn ochem are trying to get into med school, no interest at all in learning ochem itself. That means the vast majority of chemistry interns are going to be pretty useless and unmotivated in the professional labs, so the few interns trying to become actual chemists are going to get ripped off by being ignored.
Those are very chemistry specific problems but something similar might apply to many other subjects.
This will come as a surprise to the expensive equipment I used as an undergrad.
I did group lab work with an instructor as well, but my university had a pretty vibrant undergraduate research program.