Spot the Hallucination
One of these AI answers is confidently making it up. Find it.
trains the reflex the market never teaches — spotting fluent confidence as a signal independent of correctness — by making the fabrication indistinguishable from the truth until the reveal, so the miss is felt, not just described
What goes in, what comes out
A fast-paced round shows 5-8 short AI-generated answers or citations in a row, most accurate and grounded, one or two subtly fabricated — a plausible but nonexistent citation, a confidently wrong number. Learner taps the fabricated ones against the clock, and each round ends with an explainer on exactly what was faked and the one check that would have caught it.
tap every answer believed to be fabricated within the round
a precision/recall score, per-item reveal of what was fabricated and the check that would have caught it, a percentile vs other players
round length (5/8/12 items), timed vs untimed, difficulty (obvious-fake vs subtle-fake)
FREE, top-of-funnel — /playgrounds gallery, shareable quiz format for social; feeds the groundedness/verification curriculum
Concepts click when you open the machinery.
Three labs are already live and free — the same hands-on style this playground brings to its module.