A widely shared post says Webb caught a “mysterious signal” from empty space. The core is true: in the Hubble Ultra Deep Field, Webb’s NIRSpec recorded a single infrared emission line, near 1.6 micrometres, where neither Hubble nor Webb images show anything. It appeared in August and November 2023, 88 days apart, but not in October 2022 data. A distant galaxy, a supernova and a star torn apart by a black hole all fit poorly. Our take: real and unexplained, but not a message.
A post shared widely on social media this week says the James Webb Space Telescope caught a “mysterious signal” from a spot that looks completely empty, even in the deepest images ever taken. The core of it is true, with caveats. The finding comes from a peer-reviewed study in The Astrophysical Journal led by Mason Huberty of the University of Minnesota (arXiv:2610.12553). We looked at what is known and checked the claims.
In short: Webb’s spectrograph recorded one bright emission line of infrared light, at a wavelength of about 1.6 micrometres, from a point in the Hubble Ultra Deep Field where neither Hubble nor Webb images show any galaxy or star. Nobody knows yet what produced it.
FieldHubble Ultra Deep Field (Fornax)
InstrumentJWST NIRSpec, MOCSS method
Signalone emission line, ≈ 1.6 μm
SeenAug and Nov 2023, 88 days apart
Not seenJADES data, Oct 2022
StudyThe Astrophysical Journal, 2026
How the search worked, and what turned up
Original Astrometrik animation, simplified and not to scale. Left: NIRSpec’s micro-shutters close over galaxies already catalogued (dark) and stay open everywhere else, so light from uncatalogued sources can get through. Right: six spectra show the several lines of known distant galaxies; the seventh shows a single line near 1.6 μm, present in August and November 2023 but not in JADES data from October 2022.
A spectrograph turned into a search engine
NIRSpec normally points some of its quarter of a million tiny shutters at galaxies chosen in advance. Huberty’s team turned the idea round with a method they call MOCSS, multi-object coronagraphy for slitless spectroscopy: they opened almost all the shutters and closed only those over objects Hubble had already catalogued, blocking their glare. Any light that still came through could belong to something nobody had listed.
Six galaxies and one mystery
The search found seven sources with emission lines. Six matched distant galaxies already known in the field. The seventh was different: a single bright line, near 1.6 micrometres, with no second line beside it and nothing at that position in Hubble’s or Webb’s images.
It appeared in both observing windows, in August and November 2023, 88 days apart, and in all four sets of combined exposures, taken at two telescope orientations. That makes an instrument glitch unlikely. Webb’s JADES survey had observed the area in October 2022, and its data do not show the line, although it should have been bright enough. That hints the source may not have been shining then, but it is a clue, not proof.
What could it be?
One line on its own is hard to read, because we do not know how much the expansion of the universe has stretched it. If it is hydrogen’s Lyman-alpha line, the source would be a galaxy at a redshift of about 12, seen when the universe was under 400 million years old; but neutral hydrogen around such early galaxies tends to absorb exactly that line, and JADES should have seen it. A nearer galaxy or a supernova would normally show other lines as well, and they are missing. A star torn apart by a black hole can produce one strong line, but the evidence is not enough.
The authors call the object “highly unusual” and say follow-up observations are essential to pin down its distance and origin. Our take: a careful, genuine result and a good mystery, but not a message, not yet proof of a new kind of object, and not physics breaking.
Watch: the “signal from nothing” in under a minute
Our short animated video explains what Webb found in the Hubble Ultra Deep Field.
Fact check: the viral post
Claims from a widely shared post, checked:
Correct“Infrared emission from a spot that looks empty even in the deepest images”Yes: nothing appears at that position in Hubble or Webb images.
Correct“A peer-reviewed study in The Astrophysical Journal, Mason Huberty’s team, University of Minnesota”Correct (arXiv:2610.12553).
Wrong“James Webb, Hubble and JWST”James Webb and JWST are the same telescope: the images are from Hubble and Webb.
Partly right“In 2022 the spot was empty”Webb’s JADES data from October 2022 do not show the line, though it should have been bright enough. A strong hint, not proof.
Partly right“Proven not to be an error by shining steadily for 88 days”It was seen in two windows 88 days apart (August and November 2023), not watched for 88 days. Its presence in every exposure set makes an error unlikely; the authors still ask for follow-up.
Partly right“A new type of cosmic transient”Possible, but not shown: the authors list several explanations, none of which fits cleanly.
Partly right“A mysterious signal”It is light, a single emission line in a spectrum, not a radio signal or a message.
Which line could it be? Pick one
Every emission line is the fingerprint of an element, but the expanding universe stretches it to longer wavelengths. Choose which line the 1.6 μm light might be and see how far back in time the source would be.
Calculated by Astrometrik (Planck 2018 cosmology). The study itself has not identified the line.
Where is the Hubble Ultra Deep Field?
The field lies in the southern constellation Fornax, the Furnace. From Europe and Türkiye, Fornax stays low in the south on autumn and early-winter evenings. The field itself is only about a tenth of the full Moon’s width and was chosen because it has no bright stars. Our sky map shows where Fornax is from your location.
Webb’s NIRSpec recorded a single infrared emission line, near 1.6 micrometres, from a point in the Hubble Ultra Deep Field where no galaxy or star appears in Hubble or Webb images. It is light, not a radio signal or a message.
Who found it?
A team led by Mason Huberty of the University of Minnesota, in a study published in The Astrophysical Journal in 2026 (arXiv:2610.12553).
Is it a new kind of object?
Unknown. A very distant galaxy, a nearer galaxy, a supernova and a star torn apart by a black hole have been considered, and none fits cleanly. Follow-up observations are needed.
What do the 88 days mean?
The line was seen in two observing windows, in August and November 2023, which were 88 days apart. It was not watched continuously for 88 days.
Sources and calculation
Sources: M. Huberty et al., “A Single Line Emitter Identified in an Unbiased Search with JWST/NIRSpec: Multiobject Coronagraphy for Slitless Spectroscopy”, The Astrophysical Journal (2026), arXiv:2610.12553; reports by ScienceAlert (October 2026) and WION (15 October 2026). We could not open the full paper text, so details follow these reports. Redshifts and ages of the candidate lines calculated by Astrometrik with the Planck 2018 cosmology.
Positions and times were recalculated with Astronomy Engine, the sky-calculation library Astrometrik uses; times for Vienna and Istanbul are local time. The artwork on this page was drawn by Astrometrik.