AI product landscape

A map of how AI shows up in products: copilots, classifiers, recommenders, search, generative media, agents, and embedded features users barely notice.

Buying and building start with naming the job—not the logo on the API.

Ask: What input? What output? Who verifies? What’s automated vs assisted? What’s the failure cost? That sorts chat wrappers from high-stakes decision systems.

Spellcheck, photo search, fraud flags, and chat tutors are all “AI products” with different risk profiles.

List five AI features you used this week and tag each assist vs decide.

⚠️ Myth: Every product needs an agent.

✓ Reality: Many jobs want a classifier or RAG answer with citations.

⚠️ Myth: Landscape maps stay frozen.

✓ Reality: Categories shift—update yearly.

• Course 01 capabilities-and-limits; Course 25 vendors
• Course 28 curriculum maps
• NIST AI RMF: https://www.nist.gov/itl/ai-risk-management-framework ↗