How AI movie recommendations work, and when they fail

“AI movie recommendations” now means at least three different machines wearing the same label. Knowing which one you’re talking to explains both why the picks feel magical on some apps and why they feel like a horoscope on others. This is the honest tour, including the failure modes, because we build one of these things and the failure modes are where the real work is.

The three machines

• Collaborative filtering, “people who liked what you liked also liked…” This is Netflix-style recommendation and it’s genuinely powerful at scale, but it needs millions of users, drifts toward popularity, and can’t tell you why it picked anything.
• The mood quiz, most new “AI what-to-watch” apps. You type “something tense but not bleak,” a language model maps that to genres and vibes, and you get plausible titles. Fine for strangers; shallow for you, because it knows tonight’s mood but nothing about your history.
• The diary-grounded system. It measures your own ratings and other diary evidence instead of relying on what strangers liked. A language model may help express a recommendation, but a predicted star rating need not come from one. This is Kinolog’s approach, and the rest of this piece is about what makes it work or fail.

Signals aren’t equal, good systems weigh them

A rated film is stronger evidence for a personal star prediction than a quick onboarding swipe. Kinolog starts with how you use the rating scale and checks film traits against your own rated history. A watchlist save means you are interested, not that you loved the film; a never-again choice is an exclusion. When you later rate a recommended film, the stored prediction can be compared with that rating. Flattening every signal into a like loses those distinctions. See how the rating predictor works.

Notes explain context; they do not become the star formula

A note can explain why a rating mattered: “patient dread” tells a different story from a bare star. Kinolog can use that kind of private prose in optional AI recommendation features after permission, but the structured predicted-star model is a separate system built from ratings and film metadata. It does not hand your note to a language model and ask for a number. The prediction methodology shows the actual arithmetic.

Notes are still worth keeping because they make the diary more useful to you and can make conversational recommendations more specific. For a better long-term record, write the occasional real note; just do not confuse richer context with extra statistical evidence for the star predictor.

The two sins: hallucination and over-claiming

Language models invent, confidently. An AI recommender that pipes raw model output at you will eventually recommend a film that doesn’t exist, misdescribe one that does, or re-suggest something you rated last month. The fix is architectural, not hopeful: every title must be verified against a real movie database before you see it, and your library and never-again list must be enforced in code, not by politely asking the model. If an app can’t tell you it does this, assume it doesn’t.

The subtler sin is false intimacy. Three ratings in, an app announces it “knows your taste”. It doesn’t, and the picks prove it. Honest systems say so instead: early picks labeled as early reads, claims about your taste gated behind enough real data, the occasional pick that openly admits “I’ve barely seen you rate comedy. This one’s here to find out.” Distrust any recommender that’s never uncertain.

How to actually evaluate one

• Ask why. Every pick should carry a reason grounded in your history, named films, your ratings, ideally your words. “Because you like thrillers” is filler.
• Check the memory. Does it ever re-suggest what you’ve seen or sworn off? Once is a bug; twice is the architecture.
• Look for the receipts. A system confident in its picks should show its hit rate against your own verdicts, even when unflattering.
• Test the cold start. Day one, does it perform humility or fake intimacy?

That’s the standard we hold Kinolog to, TMDB-verified titles, exclusions enforced in code, reasons that cite your actual diary, and a stats page that keeps score honestly. If you try it and the picks miss, say so with one tap: correcting the machine is the fastest way to teach it.

The personalized recommendation page shows what you receive. A thin diary, a change in mood, or a film unlike your past watches can still produce a bad pick. Treat the explanation as evidence to inspect, not proof that you will enjoy the film.

Why a personal predicted rating differs from the crowd

A public score answers what a group thought; it cannot know where your tastes diverge. Our personal-vs-audience rating guideshows the evidence a credible forecast needs and why its misses matter.