I have spent a lot of time in university libraries and the disinvestment is often difficult to see unless you are doing deep research or in the weeds.
I should note that I am writing this response as somebody who is a professional researcher and scientist, not as an undergraduate student who needs to find a few references for a term paper. My needs are quite different than others and the library needs to cater to a wide range of people at all levels of research.
Many universities are in fact repurposing libraries as "study spaces" or as "learning centers". This often means dedicating more space to meeting rooms and desks, etc. To get that space, libraries either put the books in storage or quite frankly sell the books, especially books that are not checked out that often. Many of the technical books that I own are former library books. Moreover, even well-resourced government labs (like DOE national laboratories) are selling their books. I know because I have bought some (albeit through a third party). Putting books in storage is the better option, because at least the books are available, but it removes the ability to browse the shelf for adjacent books (a highly underrated research method) and often requires preparation to request books and then wait for them to arrive in a reading room. That said, it is far better to put the books in storage.
The availability of reference librarians is often greatly overstated. A library like the Library of Congress still has excellent reference librarians who often can perform apparent miracles to find difficult to locate information, but at university libraries, if I ask for a reference librarian, I often end up with an undergraduate assistant who does not really know much and cannot help you if you have a specific question to ask about locating something in a particular collection or how to use certain equipment like microfilm readers. Perhaps the undergrad can help other undergrads out, but when I was in grad school I never found this practice helpful when I needed specific information that I was already having difficulty finding.
To summarize, this trend may benefit some people, but it comes at the expense of in-depth researchers and their needs.
I will end by quibbling with that notion that "most of the information is digitized". The problem is that the Internet is so large that you can easily spend your whole life without ever realizing how shallow it really is in a single subject. Only once you go deep in that subject do you realize how much is not yet digitized in the first place. Yes, much information is digitized, but based on my experience, it is not "most". Some day that may change, but for the time being, physical libraries still play an important role here.