A galaxy confirmed this January formed just 290 million years after the Big Bang, and the part astronomers cannot explain is that it is far too large, far too soon, as if the early universe was building things it was not supposed to be able to build yet.

Light from a galaxy called JADES-GS-z14-0 left on its way to us when the universe was only about 290 million years old, and the James Webb Space Telescope has confirmed both how far away it sits and how oddly grown-up it already looks. The puzzle is not that the galaxy exists. It is that it appears far larger, far brighter and far more chemically developed than anything the models said should be possible so soon after the Big Bang.

It is worth being clear about what is surprising and what is not. The distance is a solid measurement. The maturity is the part that current theory struggles with. And none of this threatens the Big Bang itself, which rests on many separate lines of evidence. What it strains is our picture of how quickly the first galaxies could build themselves.

Looking back to 290 million years

Webb measured the galaxy’s redshift, a stretching of its light caused by the expansion of the universe, and found a value of about 14.3. That is the highest reliably confirmed for any galaxy, and it places the light we are seeing at roughly 290 million years after the Big Bang, when the cosmos was around two per cent of its present age.

The galaxy was first picked out in Webb images taken for the JWST Advanced Deep Extragalactic Survey in 2023 and 2024, then confirmed with the telescope’s spectrograph, which splits the light finely enough to nail the distance rather than merely estimate it. It became the most distant galaxy ever confirmed. Records at this frontier do not last long, and an even more remote galaxy has since been confirmed at an earlier moment still, but JADES-GS-z14-0 remains the standout case, because of what it is, not just when it is.

Too big and too bright

The strangeness starts with its size. This is not a faint speck. The galaxy spans something like 1,600 light-years and shines brightly in ultraviolet light.

The important detail is where that brightness comes from. When a distant object is unexpectedly luminous, one usual suspect is a feeding black hole, a compact point of light at the centre. Here the light is instead spread across the galaxy and looks like the combined glow of many stars. That points to a genuinely large, star-filled galaxy that had already assembled, at a time when models expected the first galaxies to be small, dim and still gathering themselves together.

The oxygen surprise

The second surprise came from a different telescope. In 2025, the ALMA array in Chile detected oxygen in the galaxy, the most distant detection of that element ever made.

Oxygen is not left over from the Big Bang. It is forged inside stars and scattered into space when they die, particularly when massive stars end as supernovae. Finding it in quantity this early means that whole generations of stars had already lived and died within those first 290 million years. The measured abundance of heavy elements ran to roughly ten times what models had predicted for such an early galaxy. An object that should have been at its very beginning already looked chemically middle-aged.

What it does and does not challenge

It is tempting to frame this as breaking physics. It is more accurate to say it is breaking a particular set of expectations. The age of the universe, its expansion history and the Big Bang framework are not in question. What is in question is the efficiency of early star and galaxy formation.

The candidate explanations all point in the same direction. Perhaps the first galaxies turned their gas into stars far more efficiently than assumed. Perhaps the feedback that normally slows star formation, the winds and blasts from young stars, was weaker in these early conditions. Perhaps the very first stars were more massive and more luminous than their modern counterparts, so a galaxy could look enormous without being quite as massive as its brightness suggests. The deepest spectroscopy yet taken of JADES-GS-z14-0 has sharpened the measurements without yet settling which of these is at work.

And this is not a story about one freak object. Since Webb began looking, it has been finding bright galaxies this early far more often than pre-Webb models predicted, by a wide margin. JADES-GS-z14-0 is the clearest single example of a pattern, which is what makes it hard to dismiss.

What to watch

Several threads will decide how big a problem this really is. One is whether these early galaxies are truly as massive as they look, or whether an unusual first generation of stars is fooling us into overestimating their mass. Another is chemistry: measuring oxygen and other elements in more galaxies this early will show whether JADES-GS-z14-0 is typical or exceptional. A third is simply the record itself, which keeps being pushed to earlier times and will test whether the pattern holds all the way back.

For now the situation is unusually honest about its own limits. The early universe appears to have built large, bright, chemically enriched galaxies faster than the textbooks allowed, the measurements have held up to repeated checking, and the explanation is still being worked out. That gap between a solid observation and a missing explanation is exactly where the interesting physics tends to live.