How machines learn
A false positive is a mistake where a model says 'yes' when the true answer is 'no', like flagging a safe email as spam or a healthy patient as sick.
Every yes-or-no prediction can go wrong in two ways. A false positive is a false alarm: the model detects something that is not there. A false negative is a miss: something real slips by undetected. Both are errors, but they hurt in different ways and are usually traded against each other.
Which error matters more depends on the situation. For cancer screening, a miss is far worse than a false alarm, so systems are tuned to flag anything suspicious. For a fraud block on your card, too many false alarms make customers furious, so the bar is set higher.
False positives are why accuracy alone is not enough. A model can be highly accurate overall while raining false alarms on a small group. Teams track the rates of both error types separately, and set the threshold based on the real cost of each mistake.
A smoke alarm shrieking because you burned toast is a false positive: it detected a fire that was not there.