LLMs don't look things up — they predict plausible-sounding text based on training data, which means they can fabricate names, dates, statistics, and citations with complete confidence. In a hiring context, this matters most when teams use AI to generate job descriptions, summarize candidate screening notes, or draft reference check questions. A hallucinated requirement, misattributed quote, or invented compliance detail can create real problems. The fix isn't to avoid AI — it's to treat its outputs as a first draft that always needs a human review before it's used.