Logging lets you refine the level of printing and is designed to make sense in the long term. There are many technical differences (structured logging, MDC etc.) that I won't get into but they are important.

To me it's mostly about the way you write the logs vs. the way you write a print. A log tries to solve the generic problem so you can deal with it in production if the need arises by dynamically enabling logs. It solves the core issue of a potential problem. A print is a local bandaid. E.g. when print debugging one would write stuff like "f*ck 1 this was reached"... In logs you would write something sensible like "Subsystem X initialized with the following arguments %s". That's a very different concept.