K2-18b, a sub-Neptune roughly 124 light-years away orbiting within its star's habitable zone, has been the subject of headline-grabbing claims about a possible biosignature. Using the James Webb Space Telescope, a University of Cambridge-led team reported in 2023 and 2025 a tentative detection of dimethyl sulfide (DMS), a gas produced predominantly by marine life on Earth, in the planet's atmosphere.
The statistical framing warrants scrutiny. The 2025 detection was reported at three-sigma significance, corresponding to about a 0.3% probability of a chance fluctuation. While that may sound compelling, the physics community conventionally reserves the language of discovery for five-sigma results, a far more demanding threshold, and three-sigma signals have historically dissipated under additional data.
Reproducibility is the deeper issue. Independent reanalyses of the same JWST observations reached different conclusions, with several finding no statistically significant evidence for DMS while confirming other species such as methane. When separate teams analysing identical data disagree, the inference is fragile and sensitive to modelling choices.
Finally, biosignature reasoning carries an abiotic-alternative problem. That a molecule is produced chiefly by life on Earth does not preclude non-biological pathways elsewhere, and the spectral signals in question are difficult to interpret and potentially confusable with other chemistry. The defensible reading is that K2-18b presents hints worth further observation, not confirmation of life, and extraordinary claims of this kind demand correspondingly strong and repeatable evidence.