K2-18b was first detected because it passes in front of its host star, causing tiny, regular dips in brightness. When the planet transits, some starlight filters through its atmosphere, imprinting spectral fingerprints that telescopes can measure. Interpreting those spectra requires careful modelling, and small differences in assumptions can change whether a signal looks like water vapour, methane, or something else.

In 2025, one analysis reported evidence for DMS or DMDS at about three-sigma significance. Other teams widened the set of molecules and atmospheric models and watched the biosignature claim vanish. The disagreement is not an empty failure of science; it is the method made visible. A spectrum does not arrive with molecule names attached. Between starlight and statement stand calibration, statistics, chemistry, and a human being who wants the answer to be life.

If K2-18b has an ocean, and if that ocean is temperate, and if chemistry there crossed into life, it happened beneath a red star in a world with no close analogue in our Solar System. Each ‘if’ matters. Held beside Earth’s stromatolites, the planet becomes neither proof nor fantasy, but a disciplined act of comparison: one biosphere close enough to touch, another possible world known only by the light that passed through its air.