arxiv: 2605.05408 · v2 · submitted 2026-05-06 · 🌌 astro-ph.IM

Recognition: no theorem link

Results of ten years of UCLA SETI searches with the Green Bank Telescope

Jean-Luc Margot

Authors on Pith no claims yet

Pith reviewed 2026-05-15 07:16 UTC · model grok-4.3

classification 🌌 astro-ph.IM

keywords SETIGreen Bank Telescopenarrowband radio signalstransmitter prevalenceanthropogenic interferenceL-band observations

0 comments

The pith

No extraterrestrial radio signals found after searching 70,000 stars

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper reports results from a decade of narrowband radio signal searches toward more than 70,000 stars and planetary systems using the Green Bank Telescope L-band receiver. A processing pipeline with 94 to 99 percent efficiency across relevant frequency drift rates identified 100 million candidate signals. Every candidate was confirmed as human-generated through automated filtering or visual inspection. With zero detections of extraterrestrial origin, the work derives a 95 percent upper limit on the fraction of stars within 20,000 light years that host a transmitter bright enough to be seen in the survey.

Core claim

After observing 70,000 stars and processing all data for narrowband drifting signals, the survey finds that every one of the 100 million candidates is anthropogenic. This null result yields a 95 percent upper bound of 6.3 times 10 to the minus 5 on the fraction of stars within 20,000 light years that host a detectable transmitter with equivalent isotropic radiated power above 5 times 10 to the 16 watts.

What carries the argument

The data-processing pipeline that achieves 94-99 percent detection efficiency for narrowband signals across drift rates of plus or minus 9 Hz per second and classifies all candidates as terrestrial or satellite interference.

If this is right

The prevalence of detectable transmitters is below 6.3 times 10 to the minus 5 at 95 percent within 20,000 light years.
The search is sensitive only to signals above an EIRP threshold of 5 times 10 to the 16 watts.
Citizen-science volunteers and AI tools are now being applied to re-examine the most interesting candidates.
The same pipeline and limits can be applied to future observations with improved receivers or larger telescopes.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The null result tightens constraints on how common long-lived, high-power radio transmitters might be in the solar neighborhood.
Extending the search to more stars or lower power levels would require either longer integration times or a larger collecting area.
The same statistical framework could be used to combine results from multiple SETI surveys to produce joint prevalence limits.

Load-bearing premise

The pipeline catches every extraterrestrial signal that could exist in the data and never mistakes one for human interference.

What would settle it

A narrowband drifting signal from one of the targeted stars that cannot be matched to any known terrestrial transmitter, satellite, or interference source.

Figures

Figures reproduced from arXiv: 2605.05408 by Jean-Luc Margot.

**Figure 1.** Figure 1: Search volume characteristics of select surveys (L-band component only), accounting for bandwidth, sky coverage, search sensitivity, pipeline efficiency, and frequency drift rate coverage. 5. Pipeline Efficiency An important, under-appreciated reality of SETI searches is that the performance of detection algorithms dramatically affects estimates of search volume (Section 4) and transmitter prevalence (Sec… view at source ↗

read the original abstract

We have been conducting a search for narrowband radio signals with the L-band receiver (1.15-1.73 GHz) of the 100 m diameter Green Bank Telescope (Margot et al., 2023). So far, we have captured radio emissions from 70,000+ stars and planetary systems in the ~9 arcminute beam of the telescope. Our data-processing pipeline has a demonstrated 94%-99% efficiency for the detection of narrowband signals across the full range of frequency drift rates (+/-9 Hz/s). All 100 million candidate signals detected to date were either automatically (99.5%) or visually (0.5%) confirmed to be anthropogenic in nature. These results allow us to place stringent limits on transmitter prevalence: at the 95% confidence level, the fraction of stars within 20,000 ly that host a transmitter that is detectable in our search (EIRP > 5e16 W) is <6.3e-5. Our most interesting signals have been uploaded to a citizen science platform (http://arewealone.earth), where 40,000+ volunteers to date have contributed insights and classifications. We are using artificial intelligence (AI) to accelerate our search, automatically excise radio frequency interference, and improve signal detection. UCLA SETI research has involved ~200 undergraduate and ~20 graduate students so far.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

This is a clean null-result update that tightens the upper limit on detectable ET transmitters from ten years of new GBT data.

read the letter

This paper reports a clean null result from ten years of SETI observations using the Green Bank Telescope's L-band receiver. The key finding is an updated upper limit: at 95% confidence, fewer than 6.3e-5 of stars within 20,000 light years host a transmitter detectable in this search, meaning EIRP greater than 5e16 W. The work is new in the sense that it incorporates fresh data from 70,000 plus targets, generating 100 million candidate signals that were all rejected as anthropogenic. The pipeline shows 94 to 99 percent efficiency across drift rates, with most rejections automatic and a small fraction visual. This supports the prevalence limit using basic statistics. It does well by documenting the long-term effort, including student training and public involvement through a citizen science platform. The approach to handling interference with AI is a practical step forward for the field. The soft spots are minor. More explicit details on how data were excluded and the exact derivation of the confidence limit would allow full verification, though the reported numbers are consistent with Poisson statistics on the effective sample. The assumption that no extraterrestrial signals were misclassified is the standard one and backed by the efficiency claims. This paper is for specialists in SETI and radio astronomy searches. It provides a useful incremental bound that refines earlier work. I would bring it to a reading group as an example of ongoing survey results. I would not cite it in my own papers in the next year, but it should go to peer review because the observations are solid and the claims are proportionate to the evidence.

Referee Report

2 major / 2 minor

Summary. The manuscript reports results from ten years of narrowband SETI observations with the Green Bank Telescope L-band receiver targeting 70,000+ stars and planetary systems. A pipeline with 94-99% demonstrated efficiency across drift rates of ±9 Hz/s identified 100 million candidates, all rejected as anthropogenic (99.5% automatic, 0.5% visual). From the null result the authors derive a 95% confidence upper limit of <6.3×10^{-5} on the fraction of stars within 20,000 ly hosting a transmitter with EIRP >5×10^{16} W. Citizen-science and AI components for future work are also described.

Significance. If the efficiency and classification claims are fully substantiated, the work supplies a well-quantified null-result constraint on transmitter prevalence from a large stellar sample, strengthening the cumulative SETI literature on the rarity of detectable technosignatures. The explicit EIRP threshold and distance limit make the bound directly usable for population studies.

major comments (2)

[Results / Upper-limit calculation] The derivation of the 95% CL limit <6.3e-5 must be shown explicitly (Poisson or equivalent formula) together with the precise effective sample size after applying the 94-99% efficiency correction to the 70,000+ targets; without this step the numerical value cannot be independently verified from the stated inputs.
[Pipeline validation] The manuscript states that all anthropogenic signals were correctly identified with no false negatives for ET-like signals, but provides no quantitative test (e.g., injection-recovery statistics) that the visual or automatic classification stage introduces zero additional false-negative rate for signals whose morphology overlaps RFI; this assumption is load-bearing for the final limit.

minor comments (2)

[Introduction / Target selection] The distance cut of 20,000 ly should be justified with a reference to the stellar catalog or distance distribution used, and the corresponding EIRP threshold should be tied to the quoted sensitivity via an explicit radiometer equation or reference.
[Figures] Figure captions and axis labels for any sensitivity or efficiency plots should include the exact frequency range, integration time, and drift-rate bins to allow direct comparison with other L-band surveys.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive review and recommendation for minor revision. We address each major comment below with clarifications and revisions to the manuscript.

read point-by-point responses

Referee: The derivation of the 95% CL limit <6.3e-5 must be shown explicitly (Poisson or equivalent formula) together with the precise effective sample size after applying the 94-99% efficiency correction to the 70,000+ targets; without this step the numerical value cannot be independently verified from the stated inputs.

Authors: We agree that an explicit derivation is required for verification. In the revised manuscript we will add a dedicated paragraph showing the Poisson statistics for a null result (upper limit λ < -ln(0.05) ≈ 3 at 95% CL) and will state the precise effective sample size N_eff after applying the measured 94-99% efficiency correction to the observed targets, from which the reported limit follows directly. revision: yes
Referee: The manuscript states that all anthropogenic signals were correctly identified with no false negatives for ET-like signals, but provides no quantitative test (e.g., injection-recovery statistics) that the visual or automatic classification stage introduces zero additional false-negative rate for signals whose morphology overlaps RFI; this assumption is load-bearing for the final limit.

Authors: The 94-99% efficiency figure already incorporates injection-recovery tests for the detection stage. For the subsequent classification of the 100 million candidates (99.5% automatic, 0.5% visual), we did not perform dedicated injection tests that quantify the false-negative rate for ET-like signals overlapping RFI morphology. We will revise the text to explicitly state this assumption, discuss its implications for the limit, and note that ongoing AI classification work will include such quantitative tests. revision: partial

Circularity Check

0 steps flagged

No significant circularity

full rationale

The central result is a standard Poisson upper limit on transmitter fraction derived from zero surviving candidates after full pipeline processing of observations toward 70,000+ stars, using a stated 94-99% detection efficiency. No parameter is fitted to the candidate data and then re-used as a prediction; the efficiency is presented as independently demonstrated rather than defined by the null result itself. The single citation to Margot et al. 2023 describes the instrument and is not load-bearing for the limit calculation. No self-definitional, fitted-input, uniqueness-imported, or ansatz-smuggled steps appear in the derivation chain.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

This is an empirical observational study. The central claim rests on standard radio astronomy assumptions about telescope sensitivity, signal classification completeness, and uniform transmitter distribution within the searched volume.

axioms (1)

domain assumption The L-band receiver observations and data-processing pipeline correctly capture and classify all narrowband signals within the stated efficiency range without systematic misses.
Invoked to support the null result and derived upper limit.

pith-pipeline@v0.9.0 · 5541 in / 1209 out tokens · 53414 ms · 2026-05-15T07:16:08.251825+00:00 · methodology

Review history (2 revisions) →

discussion (0)

Reference graph

Works this paper leans on

15 extracted references · 15 canonical work pages

[1]

Bryson, S.et al.2021, The Occurrence of Rocky Habitable-zone Planets around Solar-like Stars from Kepler Data.AJ, 161(1),

work page 2021
[2]

& Morrison, P

Cocconi, G. & Morrison, P. 1959, Searching for Interstellar Communications.Nature, 184, 844–846. Cordes, J. M. & Lazio, T. J. 1991, Interstellar Scattering Effects on the Detection of Narrow-Band Signals. ApJ, 376,

work page 1959
[3]

1984,.SETI Science Working Group Report, chapter Estimates of the Relative Probability of Success of the SETI Search Program, pp

Drake, F. 1984,.SETI Science Working Group Report, chapter Estimates of the Relative Probability of Success of the SETI Search Program, pp. 67–69. Number TP-2244 in NASA Technical Paper. Drake, F. D. 1961, Project Ozma.Physics Today, 14, 40–46. Dreier, C. 2025,. Your Guide to NASA’s Budget.https://www.planetary.org/space-policy/ nasa-budget. Enriquez, J. ...

work page 1984
[4]

Gaia Collaboration 2023, Gaia Data Release

work page 2023
[5]

Gajjar, V

Summary of the content and survey properties.A&A, 674, A1. Gajjar, V . & Brown, G. C. 2026, Exo-IPM Scattering as a Hidden Gatekeeper of Narrowband Technosignatures.ApJ, 999(2),

work page 2026
[6]

R.et al.2016, SETI Observations of Exoplanets with the Allen Telescope Array.AJ, 152,

Harp, G. R.et al.2016, SETI Observations of Exoplanets with the Allen Telescope Array.AJ, 152,

work page 2016
[7]

& Sagan, C

Horowitz, P. & Sagan, C. 1993, Five years of Project META - an all-sky narrow-band radio search for extraterrestrial signals.ApJ, 415, 218–235. Koron, A. & Margot, J. L. 2026, A deep learning search for technosignatures.In preparation,. Lesyna, L. 2019, A Technosignature Carrying a Message Will Likely Inform us of Crucial Biological Details of Life Outsid...

work page 1993
[8]

Li, M. G.et al.2026, Are we alone in the universe? - a citizen science collaboration and machine learning approach to mitigating radio frequency interference.Astronomical Journal, Submitted,. Ma, P. X.et al.2023, A deep-learning search for technosignatures from 820 nearby stars.Nature Astronomy, 7, 492–502. Margot, J. L., Croft, S., Lazio, J., Tarter, J.,...

work page 2026
[9]

L.et al.2018, A Search for Technosignatures from 14 Planetary Systems in the Kepler Field with the Green Bank Telescope at 1.15–1.73 GHz.AJ, 155,

Margot, J. L.et al.2018, A Search for Technosignatures from 14 Planetary Systems in the Kepler Field with the Green Bank Telescope at 1.15–1.73 GHz.AJ, 155,

work page 2018
[10]

Margot, J.-L.et al.2021, A Search for Technosignatures around 31 Sun-like Stars with the Green Bank Telescope at 1.15-1.73 GHz.AJ, 161(2),

work page 2021
[11]

L.et al.2023, A Search for Technosignatures Around 11,680 Stars with the Green Bank Telescope at 1.15–1.73 GHz.AJ, 166(5),

Margot, J. L.et al.2023, A Search for Technosignatures Around 11,680 Stars with the Green Bank Telescope at 1.15–1.73 GHz.AJ, 166(5),

work page 2023
[12]

L., Greenberg, A

Margot, J. L., Greenberg, A. H., Pinchuk, P.,et al.2020,. Dryad Data Set. https://doi.org/10.5068/D1309D. M´endez, A.et al.2025, Arecibo Wow! II: Revised Properties of the Wow! Signal from Archival Ohio SETI Data.arXiv e-prints,, arXiv:2508.10657. Messerschmitt, D. G. 2013, End-to-end interstellar communication system design for power efficiency. arXiv e-...

work page doi:10.5068/d1309d 2020
[13]

& Margot, J

Pinchuk, P. & Margot, J. L. 2022, A Machine Learning-based Direction-of-origin Filter for the Identification of Radio Frequency Interference in the Search for Technosignatures.AJ, 163(2),

work page 2022
[14]

C.et al.2020, The Breakthrough Listen Search for Intelligent Life: Observations of 1327 Nearby Stars Over 1.10–3.45 GHz.AJ, 159(3),

Price, D. C.et al.2020, The Breakthrough Listen Search for Intelligent Life: Observations of 1327 Nearby Stars Over 1.10–3.45 GHz.AJ, 159(3),

work page 2020
[15]

2001, The Search for Extraterrestrial Intelligence (SETI).Annual Review of Astronomy and Astrophysics, 39, 511–548

Tarter, J. 2001, The Search for Extraterrestrial Intelligence (SETI).Annual Review of Astronomy and Astrophysics, 39, 511–548. Wlodarczyk-Sroka, B. S., Garrett, M. A., & Siemion, A. P. V . 2020, Extending the Breakthrough Listen nearby star survey to other stellar objects in the field.MNRAS, 498(4), 5720–5729. Wright, J. 2025,. NSF and NASA Funding for SE...

work page 2001