ClearFactAll topics

Is terrestrial life likely common on exoplanets?

3 peer-reviewed studies for, 4 against

Key takeaways

Stronger evidence: Against

Verdict

Whether terrestrial life is likely common on exoplanets is a question with two distinct empirical layers: how frequently do rocky habitable-zone planets exist, and how reliably does life arise on them. The "for" side has strong evidence for the first layer and tentative evidence for the second. Bryson, Kunimoto and colleagues (Astronomical Journal, 2021) used the final Kepler DR25 catalogue with Gaia stellar revisions to estimate η⊕ = 0.37–0.88 planets per Sun-like star, putting hundreds of millions of rocky habitable-zone planets in the Milky Way around Sun-like stars alone. Dressing and Charbonneau (Astrophysical Journal, 2015) showed that M dwarfs — by far the most common stellar type — host roughly 2.5 small planets each at radii 1–4 R⊕ and periods shorter than 200 days, with a substantial fraction in the habitable zone, implying tens of billions of small potentially habitable planets in the galaxy. Madhusudhan and colleagues (Astrophysical Journal Letters, 2023) detected CH₄ and CO₂ in the atmosphere of K2-18 b, a sub-Neptune in its star's habitable zone, demonstrating that JWST can characterise such atmospheres at the sensitivity required to test biosignature hypotheses.

The "against" side concedes the planet-frequency claim and concentrates on the much harder claim that life is therefore common. Catling, Krissansen-Totton and colleagues (Astrobiology, 2018) developed the field-standard framework for biosignature assessment and catalogued the many abiotic processes — photochemical O₂, serpentinisation-driven CH₄, and others — that can mimic biological signals, concluding that no single molecular detection rises to the level of evidence for life without contextual information about planetary geochemistry and stellar environment. Lingam and Loeb (Reviews of Modern Physics, 2019) reviewed the physical factors regulating low-mass-star habitability and argued that intense stellar winds, frequent superflares, tidal locking, and atmospheric escape may render many of the most numerous candidate planets biologically uninhabitable in ways that standard habitable-zone definitions do not capture. Luque and colleagues (Astronomy & Astrophysics, 2025) re-analysed the full panchromatic JWST spectrum of K2-18 b and found insufficient evidence for the DMS/DMDS biosignature claim — alternative non-biological molecules fit the data equally well, and confident detection would require roughly 26 additional MIRI transits.

The synthesis: rocky habitable-zone planets are unambiguously common; life on them is unambiguously not yet detected. The "for" side's strongest claim — that terrestrial-style biospheres should be common because Earth-like substrates are common — relies on a translation from physical occurrence to biological occurrence that the empirical record does not yet support and that the "against" papers identify as substantially constrained by stellar-environment physics, abiotic-mimic chemistry, and rigorous detection thresholds. The "against" side prevails because the question, as posed, is whether life is likely common on exoplanets, and the honest scientific answer is that we currently have zero confirmed extraterrestrial biospheres, one heavily disputed candidate biosignature, well-developed reasons to doubt that all rocky habitable-zone worlds host life, and a framework that explicitly raises the evidentiary bar above the level any current observation has cleared. The for-side papers correctly establish the prior — habitable-zone real estate is abundant — but the leap from prior to posterior requires evidence that does not currently exist. The right scientific posture is curiosity and patience, not confidence that terrestrial life is common.

Strongest study on each side

Ranked by citation count.

Strongest study for

Carbon-bearing Molecules in a Possible Hycean Atmosphere (Astrophysical Journal Letters, 2023)

Nikku Madhusudhan, Subhajit Sarkar et al.

Madhusudhan, Sarkar, Constantinou, Holmberg, Piette, and Moses reported JWST NIRISS and NIRSpec observations of K2-18 b, a temperate sub-Neptune in the habitable zone of an M dwarf, and detected methane (CH₄) and carbon dioxide (CO₂) at 5σ and 3σ confidence with volume mixing ratios of order 1% in a hydrogen-rich atmosphere. The non-detection of NH₃ alongside abundant CH₄ and CO₂ is consistent with predictions for a "hycean" world — a temperate hydrogen atmosphere overlying a global ocean. The authors also reported a tentative DMS feature, motivating consideration of biological activity. Whether or not the DMS hint survives, the result demonstrates that exoplanetary atmospheres are now spectroscopically accessible at the sensitivity required to test biosignature hypotheses on individual targets.

Strongest study against

Open the interactive page →

The evidence

For (3)

Carbon-bearing Molecules in a Possible Hycean Atmosphere (Astrophysical Journal Letters, 2023)

Nikku Madhusudhan, Subhajit Sarkar et al.

Madhusudhan, Sarkar, Constantinou, Holmberg, Piette, and Moses reported JWST NIRISS and NIRSpec observations of K2-18 b, a temperate sub-Neptune in the habitable zone of an M dwarf, and detected methane (CH₄) and carbon dioxide (CO₂) at 5σ and 3σ confidence with volume mixing ratios of order 1% in a hydrogen-rich atmosphere. The non-detection of NH₃ alongside abundant CH₄ and CO₂ is consistent with predictions for a "hycean" world — a temperate hydrogen atmosphere overlying a global ocean. The authors also reported a tentative DMS feature, motivating consideration of biological activity. Whether or not the DMS hint survives, the result demonstrates that exoplanetary atmospheres are now spectroscopically accessible at the sensitivity required to test biosignature hypotheses on individual targets.

The Occurrence of Rocky Habitable-zone Planets around Solar-like Stars from Kepler Data (Astronomical Journal, 2021)

Steve Bryson, Michelle Kunimoto et al.

Bryson, Kunimoto, Kopparapu, and colleagues used the final Kepler DR25 catalogue with revised stellar parameters from Gaia DR2 to estimate η⊕, the average number of rocky habitable-zone planets per Sun-like star. They found η⊕ = 0.37–0.60 planets per star for the conservative habitable zone and 0.58–0.88 for the optimistic habitable zone. Even using the conservative end, this implies hundreds of millions of rocky habitable-zone planets in the Milky Way around Sun-like stars alone — a robust empirical anchor for the claim that the physical conditions plausibly required for terrestrial life are not rare in the galaxy.

The Occurrence of Potentially Habitable Planets Orbiting M Dwarfs Estimated from the Full Kepler Dataset (Astrophysical Journal, 2015)

Courtney D. Dressing and David Charbonneau

Dressing and Charbonneau analysed the full Kepler M-dwarf sample with empirical detection-sensitivity injections and estimated 0.56 Earth-sized planets and 0.46 super-Earths per M dwarf at periods shorter than 50 days, with a cumulative occurrence of 2.5 ± 0.2 planets per M dwarf at radii 1–4 R⊕ and periods shorter than 200 days. Within the conservative habitable zone, they estimated occurrence rates that imply nearby M-dwarf habitable-zone planets should exist within roughly 5 parsecs of Earth. Because M dwarfs are by far the most common stars in the galaxy, this places the total number of small, potentially habitable planets in the Milky Way in the tens of billions.

Against (4)

Insufficient evidence for DMS and DMDS in the atmosphere of K2-18 b: a joint analysis of JWST NIRISS, NIRSpec, and MIRI observations (Astronomy & Astrophysics, 2025)

Luque, Piaulet-Ghorayeb, Radica, Xue, Zhang, Bean, Samra, and Steinrueck performed the first joint analysis of the panchromatic 0.6–12 μm JWST spectrum of K2-18 b, combining NIRISS, NIRSpec, and MIRI observations and testing alternative atmospheric models and data reductions. They concluded there is insufficient evidence for the presence of dimethyl sulfide or dimethyl disulfide in the atmosphere — alternative methyl-bearing molecules such as ethane fit the data equally well, and the marginal preferences seen in earlier analyses depend sensitively on modelling choices. They estimated that roughly 26 additional MIRI transits would be required to achieve a statistically significant detection. The result is a direct empirical caution against treating the most prominent recent exoplanet "biosignature" claim as evidence of life.

Exoplanet Biosignatures: A Framework for Their Assessment (Astrobiology, 2018)

David C. Catling, Joshua Krissansen-Totton et al.

Catling, Krissansen-Totton, Kiang, Crisp, Robinson, DasSarma, Rushby, Del Genio, Bains, and Domagal-Goldman developed the foundational framework for evaluating proposed exoplanet biosignatures. The paper systematically catalogues abiotic processes that produce false-positive biosignatures — including photochemical O₂ from H₂O photolysis, abiotic CH₄ from serpentinisation, and many others — and develops a Bayesian framework requiring that planetary context (atmospheric composition, stellar UV, geochemistry) be evaluated alongside the candidate biosignature. The framework's conclusion is that no single molecular detection should be taken as evidence of life; the bar for confidently inferring biology from a remote spectrum is much higher than enthusiastic individual detections suggest.

Colloquium: Physical constraints for the evolution of life on exoplanets (Reviews of Modern Physics, 2019)

Manasvi Lingam and Abraham Loeb

Lingam and Loeb reviewed the physical factors that regulate whether worlds — particularly those around low-mass stars where most rocky habitable-zone planets are expected — can plausibly host complex biospheres. They argued that several mechanisms place serious constraints on M-dwarf habitability: intense stellar winds and XUV flux causing rapid atmospheric escape, frequent superflares delivering harmful UV doses, tidal locking concentrating biological activity, and an ocean–continent fraction that may not be conducive to land life. The colloquium concludes that the high physical-occurrence rate of rocky habitable-zone planets does not translate directly into a high frequency of life — many of the most numerous candidate planets may be biologically uninhabitable for stellar-environment reasons that are not captured in standard habitable-zone definitions.

The debate

Maya: Start with the planet-occurrence numbers because they are extraordinary. Bryson, Kunimoto and colleagues used the final Kepler DR25 catalogue with Gaia stellar revisions and pinned η⊕ — the average number of rocky habitable-zone planets per Sun-like star — at between 0.37 and 0.88. Dressing and Charbonneau showed that M dwarfs, the most common stellar type, host roughly 2.5 small planets each. Together these imply tens of billions of small habitable-zone planets in the Milky Way alone. The substrate for terrestrial life is everywhere.

Daniel: I will agree completely with the planet count. The substrate is abundant. The question on the table is not whether rocky habitable-zone planets exist — it is whether terrestrial life is therefore likely common on them. Those are different claims. Lingam and Loeb in their 2019 colloquium in Reviews of Modern Physics argued that around low-mass stars, where most of those rocky habitable-zone planets sit, intense stellar winds, frequent superflares, atmospheric escape and tidal locking severely constrain habitability in ways that the standard habitable zone definition does not capture. Many of the most numerous candidate planets may be biologically uninhabitable for reasons that have nothing to do with the equilibrium temperature.

Maya: Those constraints are real but their severity is contested in the literature. We do not actually know yet how often M-dwarf rocky planets retain atmospheres, and the JWST era is just beginning to test those predictions. And on the positive end of the evidence, Madhusudhan and colleagues in 2023 in the Astrophysical Journal Letters reported the detection of methane and carbon dioxide in the atmosphere of K2-18 b, a sub-Neptune in the habitable zone of an M dwarf, with mixing ratios of order 1% — consistent with predictions for a hycean ocean world. They also reported a tentative DMS detection that, if confirmed, would be a strong biosignature.

Daniel: The "if confirmed" is doing enormous work in that sentence. Luque, Piaulet-Ghorayeb and colleagues did exactly the confirmatory analysis in Astronomy & Astrophysics in 2025 — joint analysis of NIRISS, NIRSpec, and MIRI observations across the full 0.6 to 12 micron range. They found insufficient evidence for DMS or DMDS. Alternative non-biological molecules like ethane fit the data equally well. The marginal preferences for DMS in earlier analyses are sensitive to modelling choices, and a confident detection would need roughly 26 additional MIRI transits. The most prominent recent exoplanet biosignature candidate has not held up.

Maya: That is a real result and I take it seriously. But it is one target. The broader argument is not that K2-18 b specifically hosts life — it is that the era of remote spectroscopic life-detection has begun. JWST is now the right tool for the right targets, and the next decade will produce many such tests. The for-side argument is statistical: with tens of billions of habitable-zone planets and improving detection capability, the probability that none host detectable biospheres is low under reasonable abiogenesis priors.

Daniel: That argument does the same translation I keep flagging. It substitutes a planet-frequency claim for a life-frequency claim. We have one example of life — Earth — and one example of abiogenesis — Earth. That is not a sample we can use to set a confident prior on how often life arises given the right substrate. Catling, Krissansen-Totton and colleagues in 2018 in Astrobiology made exactly this point. The bar for confidently inferring biology from a remote spectrum is high, because abiotic chemistry can mimic almost every individual signal we hope to use. Without contextual information about planetary geochemistry and stellar environment, no single detection settles the question.

Maya: I agree the bar is high. The point of building JWST and the next-generation observatories — habitable-worlds telescopes, ground-based ELTs — is to clear that bar by accumulating contextual evidence on individual targets. The Catling framework is the playbook for how to do that. It is not an argument against the program; it is an argument for doing the program rigorously.

Daniel: Agreed on the program. What I push back on is the framing of the question. "Is terrestrial life likely common on exoplanets?" is currently answered by the empirical record as: we do not know, and the only positive signal we had has been retracted on closer analysis. Saying yes here would be doing exactly what the Catling framework warns against — taking the prior as evidence and skipping the posterior. Saying no would be overstepping in the other direction, but it is closer to the data.

Maya: I would put it as: the prior is favourable, the technology to update the prior is now operational, and the right scientific posture is empirical patience, not confident yes-or-no. But asked to choose, I would say the for side has stronger grounds because the planet-frequency case is solid and history shows that nature tends to fill possible niches when the chemistry permits. The against side's strongest arguments are about evidentiary discipline, not about reasons to expect Earth to be alone.

Daniel: That is the right frame, and we agree on the empirical patience. Where we disagree is whether "the prior is favourable" justifies the conclusion that life is likely common. The Catling, Lingam-Loeb, and Luque papers all argue, in different ways, that the gap between "habitable substrate" and "biosphere" is wider than the planet-frequency papers alone suggest. The honest answer to the topic question, on present evidence, is closer to no than to yes.

Frequently asked questions

Have scientists found life on any exoplanet?

No. There are currently no confirmed extraterrestrial biospheres, and the most prominent candidate biosignature, on K2-18 b, is heavily disputed.

How many potentially habitable planets are in the Milky Way?

Kepler-based estimates suggest hundreds of millions of rocky habitable-zone planets around Sun-like stars alone. Including M dwarfs, the number of small, potentially habitable planets may reach tens of billions.

Did JWST detect signs of life on K2-18 b?

JWST detected methane and carbon dioxide in the atmosphere of K2-18 b, along with a tentative dimethyl sulfide (DMS) feature. A later joint analysis of the full spectrum found insufficient evidence for DMS or DMDS, since non-biological molecules fit the data equally well. That analysis estimated that roughly 26 more MIRI transits would be needed for a confident detection.

Why doesn't a planet in the habitable zone mean it has life?

Standard habitable-zone definitions do not account for factors such as intense stellar winds, superflares, atmospheric escape and tidal locking, which may make many such planets uninhabitable. The existence of suitable planets also says nothing directly about how often life actually arises on them.

Related questions