Serif vs sans-serif
Serifs are the little feet on the ends of strokes; sans-serif fonts have none. Feet read bookish, no feet reads clean and modern.
See it
What it is
A serif is the small stroke finishing the end of a letter's arm or stem: the feet on Times New Roman, Georgia, Garamond. Sans-serif means without them: Helvetica, Inter, Arial. That is the whole distinction. Everything else people attach to it (serif is formal, sans is modern) is convention, not physics, and plenty of designs invert it deliberately.
Practically, serifs read as editorial, literary, institutional, trustworthy-old-money. Sans reads as neutral, technical, product-y, Swiss. The classic pairing move is one of each: a serif for headlines with a sans for body, or the reverse for a magazine feel. Slab serifs (Roboto Slab, Rockwell) sit in the middle with blocky rectangular feet and read sturdy rather than refined.
Gotcha: the old claim that serifs are more readable in print and sans on screen was a hardware artifact. Low-resolution displays mangled fine serifs; retina screens do not, and serif body copy on the web now reads fine. What still holds is that delicate high-contrast serifs (Didot, Bodoni) fall apart at small sizes and in light weights, so keep those for display and pick a sturdy text serif for paragraphs.
Ask AI for it
Set this landing page with a serif and sans pairing: headlines in a high-contrast display serif (Playfair Display) at weight 700 with tight tracking, body copy in a humanist sans (Source Sans 3 or Myriad) at 400 with line-height 1.6 and a 65ch measure. Keep labels, buttons, and small UI text in the sans so nothing under 14px relies on fine serifs.