Astronomers studying the exoplanet K2-18 b have found that earlier hints of dimethyl sulfide and dimethyl disulfide, two molecules tied to biological activity on Earth, do not hold up when data from multiple James Webb Space Telescope instruments are analyzed together. The finding weakens one of the most talked-about possible biosignature claims in recent exoplanet science and raises hard questions about how telescope observations should be processed before any life-related announcement is made.
Why the K2-18 b biosignature claim matters right now
K2-18 b sits in the habitable zone of its host star and has drawn intense attention since JWST observations first suggested its atmosphere might contain gases associated with living organisms. A mid-infrared study argued that the planet’s spectrum was difficult to explain with many candidate molecules except DMS and DMDS, feeding public excitement about a possible life signal. That excitement, though, rested on data from a single instrument and a single data-reduction framework, a narrow foundation for such a large claim.
The central question is straightforward: when researchers combine all publicly available JWST datasets for K2-18 b and run them through consistent, community-standard methods, does the statistical support for DMS or DMDS survive? The emerging answer, drawn from several independent reanalyses, is that it does not. When near-infrared data from NIRISS and NIRSpec are jointly analyzed alongside mid-infrared MIRI observations, the Bayesian evidence ratio for these molecules drops below the threshold that scientists typically require to call something a detection.
Joint JWST reanalysis erodes DMS and DMDS support
A joint analysis combining JWST NIRISS, NIRSpec, and MIRI observations of K2-18 b found insufficient evidence for DMS and DMDS in the planet’s atmosphere. The study tested whether the earlier mid-infrared result held when all available wavelength ranges were included and when different binning choices were applied to the raw data. It did not. The signal that appeared under one particular reduction approach faded when the data were handled differently or when additional spectral coverage was folded in.
A separate assessment went further, concluding that K2-18 b does not meet the standards of evidence for life. That analysis focused on how sensitive the claimed detections were to the specific way observations were binned, finding that most retrievals under alternative binning schemes did not support the presence of DMS or DMDS. The implication is clear: the original result depended heavily on choices made during data processing rather than on a strong, repeatable atmospheric signal.
A peer-reviewed study published in The Astrophysical Journal Letters in 2026 systematically searched for trace molecules in K2-18 b’s atmosphere and reported that only marginal support for certain species remained, and only under narrow assumptions about how the data were reduced. That work reinforced the pattern seen across multiple independent efforts: the more carefully the data are treated, the weaker the biosignature case becomes.
An earlier reanalysis of the JWST NIRISS and NIRSpec transmission spectrum had already flagged concerns, noting that notable molecule claims were derived using a single reduction and retrieval framework without standard robustness tests. That study questioned whether the original results could be reproduced and called for the kind of multi-instrument, multi-pipeline approach that the later joint analysis ultimately carried out.
What NASA says and what the data still lack
NASA’s own science explainer on exoplanet biosignatures lists DMS as a possible indicator of life and references the K2-18 b findings, but the agency’s framing is cautious. The page distinguishes between a “strong indication” and a confirmed discovery, and the K2-18 b results fall well short of the latter category. No official NASA statement has directly addressed the 2025 and 2026 reanalyses that have weakened the original claim.
Several gaps in the evidence remain open. No direct quotes from the lead authors of the joint analysis or the standards-of-evidence assessment have been published in the available record, making it difficult to gauge how those teams interpret the broader implications of their work. The raw JWST observation identifiers and exact binning parameters used across the various studies have not been fully cataloged in a single, publicly accessible comparison, which limits the ability of outside researchers to replicate every step.
The peer-reviewed status of some key papers also deserves attention. While the systematic trace-molecule search appeared in The Astrophysical Journal Letters, several of the reanalyses that form the backbone of the counter-argument remain as preprints on arXiv. Peer review does not guarantee correctness, but it does subject methods and conclusions to formal scrutiny that preprints have not yet received. Until more of these studies clear that hurdle, some astronomers will reserve judgment on the final word about K2-18 b’s atmosphere.
At the same time, the lack of strong evidence is not evidence of absence. JWST has only observed a handful of K2-18 b transits and eclipses so far, and the signal-to-noise ratio for subtle spectral features remains modest. Clouds or hazes in the planet’s atmosphere could be muting or masking signatures of minor species, including any potential biosignature gases. It is also possible that the chemistry of a sub-Neptune in the habitable zone does not map cleanly onto Earth-based expectations, complicating the interpretation of any marginal features that do appear.
Upcoming JWST cycles will shape the next chapter for K2-18 b
The practical consequence of these findings extends beyond a single planet. JWST observing time is a finite and heavily competed resource. If the biosignature case for K2-18 b continues to weaken, telescope allocation committees will face pressure to redirect future cycles toward other targets where atmospheric signals are less ambiguous and where retrievals are less sensitive to data-handling choices. That could mean prioritizing smaller, more Earth-sized worlds around very quiet stars, or gas-rich planets whose spectra are easier to interpret.
Still, K2-18 b is unlikely to vanish from the schedule altogether. The planet remains one of the few known sub-Neptunes in a temperate orbit with high-quality JWST spectra across multiple instruments. That makes it a valuable benchmark for understanding a class of worlds that appears common in the galaxy but has no direct analogue in our own solar system. Even if DMS and DMDS are not robustly detected, refining the planet’s bulk composition, temperature profile, and cloud structure will help calibrate models used for many other exoplanets.
Future JWST cycles are expected to bring deeper observations that could clarify some of the current ambiguities. Longer integrations would improve the precision of transmission spectra, especially in wavelength regions where potential features from trace gases overlap. Additional instrument modes could probe different altitudes in the atmosphere, testing whether any weak signals are confined to particular pressure levels. Coordinated campaigns with ground-based observatories might also help track stellar variability that could be contaminating the spectra.
On the analysis side, astronomers are pushing toward more transparent and standardized pipelines. Several teams are now publishing not just their final spectra but also intermediate data products and full retrieval configurations, allowing others to rerun and modify their workflows. For contentious cases like K2-18 b, that level of openness may become a de facto requirement before journals or agencies treat any biosignature claim as credible.
The K2-18 b episode is already shaping how the community talks about life beyond Earth. The rapid rise and partial fall of the DMS and DMDS hints underscore the need for conservative language, multiple independent confirmations, and clear statistical thresholds before anything is framed as evidence for biology. They also highlight the power-and the peril-of revolutionary instruments like JWST: when the data are unprecedented, the temptation to over-interpret them can be strong.
For now, K2-18 b stands as a cautionary tale rather than a herald of alien life. The planet remains intriguing, its atmosphere still only partially understood, but the specific biosignature claims that once made headlines have lost much of their support under closer scrutiny. As new observations arrive and methods improve, the story may evolve again. Whatever the outcome, the lessons learned from this world will influence how scientists search for and evaluate possible signs of life across the cosmos.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
