Webb Found Brown Dwarfs With Only Two Jupiter Masses

The image looks like a wall of dust lit by young stars, but the strangest part is almost hidden in the glow. On September 15, 2026, NASA and ESA released one of the largest James Webb Space Telescope views yet of IC 348, a star-forming region about 1,000 light-years away in the constellation Perseus. The beautiful detail is the nebula. The uncomfortable detail is something else: Webb found brown dwarfs with only two Jupiter masses.

According to NASA's report and ESA's release, these are the lowest-mass objects of their kind yet known. They are not normal stars, because they never become hot enough to sustain hydrogen fusion. They are not planets in the comfortable sense either, because they appear to form directly from collapsing molecular clouds, the same process that creates stars. They sit in a boundary zone where our language starts to fail.

The lower limit became less comfortable

For a long time, brown dwarfs were described as "failed stars". The phrase is useful, but it simplifies too much. They form like stars, through gravitational collapse, and many can briefly burn deuterium early in life. What they cannot do is sustain the hydrogen fusion that keeps a star alive for millions or billions of years. Below roughly 8% of the Sun's mass, we enter that brown, cool territory.

The problem is that two Jupiter masses look too small for that process, at least under many models. The team had already used Webb in 2022 to find objects of three to four Jupiter masses in IC 348. Now, with NIRCam data collected in 2024 and NIRSpec spectroscopy in 2025, it has pushed even lower. NASA notes that two Jupiter masses equal only 0.19% of the Sun's mass. That is not only a record; it is a new question about how far a cloud can fragment before it stops producing star-like objects.

When a planet-sized object is born like a star

The discovery matters because it mixes two stories we usually keep separate. In a classic planetary system, planets grow inside a disk of material around a star. In IC 348, Webb is suggesting that planetary-mass objects can form in isolation, as miniatures of the stellar process. One of the lightest objects even shows signs of a disk, raising an almost dizzying possibility: small planets forming around something that, by mass, already resembles a planet.

This does not mean Jupiter was an aborted star, or that all giant planets should be reclassified. It means nature may use similar recipes at different scales. The border between giant planet, brown dwarf, and protostar stops being a clean line and becomes a transition zone where origin, mass, and environment all matter at once.

Webb is turning nebulae into population laboratories

The strength of this observation is not sensitivity alone. It is the combination of image and spectrum. First, NIRCam identifies candidates by infrared colour and brightness. Then NIRSpec helps estimate masses and atmospheric properties. While examining the spectra, the team also found a feature attributed to hydrocarbons, molecules made only of carbon and hydrogen, seen in the lowest-mass objects. ESA notes that this signature may point to a spectral class of its own for these extremes.

IC 348 also offers context. In the upper-right corner of the image, protostars launch jets into the surrounding gas and dust, creating Herbig-Haro objects such as HH 797 and HH 211. The same picture therefore shows the full theatre: stars being born, jets sculpting clouds, brown dwarfs challenging minimum masses, and even background galaxies cutting through the composition. Webb is not merely taking portraits; it is counting how many different forms stellar birth can assume.

Why it matters now

The news arrives at a point where Webb is no longer only a machine for spectacular images. Its scientific value is appearing at the limits: how early galaxies form, which molecules exist in distant atmospheres, how light objects born like stars can be. IC 348 fits exactly into that shift because it turns a nearby nebula into a direct test of molecular-cloud fragmentation models.

If these objects are confirmed as part of a common population, we will have to explain why star formation can manufacture almost planetary bodies without needing a parent star. If they are rare, they still mark the extreme boundary of a process we do not yet fully describe. Either way, the result is useful: it shows that the universe keeps producing exceptions small enough to fit in Jupiter's mass range, but large enough to force us to rewrite our practical definition of stellar birth.