Every time a new technology touches libraries, the same question resurfaces: does this replace the librarian, or does it hand the librarian a better tool? AI is no exception, and the honest answer is a bit of both, depending on the task.
Search and discovery are the clearest wins. AI-assisted search can help a user go from a vague question to a relevant source far faster than browsing a catalog alone. Cataloguing and metadata work benefit too — tasks that used to take hours of manual tagging can be drafted in minutes and then reviewed by a human who understands the collection.
Where AI struggles is judgment: knowing which source is trustworthy, understanding a community's specific needs, deciding what belongs in a curated collection versus what's just noise. That's still deeply human work, and it's where librarians' expertise becomes more valuable, not less.
The opportunity, then, isn't in resisting these tools or blindly adopting them — it's in figuring out which parts of the job get faster, so more time goes toward the parts that were always the real value: guiding people to the right information and helping them make sense of it.