Karl Popper proposed falsifiability as a demarcation criterion for distinguishing science from non-science. A scientific claim, in his view, must make predictions that could in principle conflict with possible observations.
Popper developed this idea in response to the problem of induction: no number of confirming observations can logically prove a universal law. Seeing thousands of white swans cannot establish that every swan is white, but one genuine black swan can refute that claim. Falsifiability therefore emphasizes risky tests and potential counterexamples rather than final verification.
Popper introduced the idea in The Logic of Scientific Discovery, first published in German in 1934 and later translated into English. He used it to criticize theories such as some forms of psychoanalysis and Marxism, which he argued could accommodate almost any possible observation.
A common mix-up is confusing falsifiability with falsification. Falsifiability means that a claim is testable in a way that could disprove it; it does not mean the claim has actually been disproved. Later philosophers, including Thomas Kuhn and Imre Lakatos, criticized or refined aspects of Popper's account.