The half-life problem

Specific technical content dates, and in fast-moving fields it dates quickly — the particular framework, the current regulation, the tool everyone uses this year. Some of what you are taught in first year will be wrong or obsolete before you graduate, and every honest lecturer knows it.

That is not an argument against studying it. You cannot learn to think in a field without learning the field as it currently stands. It is an argument against measuring the degree by how much of the syllabus you can still recite in five years, because the answer will be almost none of it and that is not the failure it looks like.