In 2025, a paper from a South Korean research team drew attention across astronomy. It suggested that the evidence for dark energy itself might be wrong. Dark energy is thought to make up about 70% of the universe, yet no one knows what it is, which makes it a natural target for skeptical scrutiny.
Scientists have long known that the universe is expanding. In the late 1990s, however, it was discovered that this expansion is accelerating. Astrophysicists concluded that something must be driving the acceleration, and they named this unknown ingredient "dark energy."
The 2025 re-analysis, by a team at Yonsei University in Seoul, argued that the expansion of the universe is not actually accelerating.
Our team at the University of Southampton re-examined the Yonsei team's results. Our findings, published in the journal Monthly Notices of the Royal Astronomical Society (MNRAS), suggest that the 2025 result was mistaken.
The Yonsei team, however, stands by its findings. It says that a follow-up study also supports the possibility that the universe is not accelerating.
What is dark energy?
In the late 1990s, cosmology was upended by a startling discovery. Scientists knew the universe was expanding, but they believed gravity was gradually slowing that expansion.
But by observing extremely bright events, the thermonuclear explosions of white dwarfs, two independent teams of astronomers found the opposite. These explosions, known as Type Ia supernovae, have nearly the same intrinsic brightness, so comparing their apparent brightness allows precise distance measurements.
In 1998, the data revealed that distant supernovae were fainter than expected, dimmer than they should be in a decelerating universe. Because faintness implies greater distance, the observations showed that these supernovae were farther away than predicted, suggesting that cosmic expansion was not slowing down.
When the calculations were actually done, they showed that the expansion was in fact accelerating.
No convincing physical theory existed to explain this "accelerating expansion of the universe." Physicists gave the mysterious repulsive force behind the acceleration the name "dark energy."
Calibrating the universe
To understand the Yonsei group's claim, you need to understand how astronomers use Type Ia supernovae as "standard candles" to measure cosmic distances. Type Ia supernovae are very similar to one another, but not perfectly identical.
Cosmologists correct for these differences. They apply subtle calibrations based on the relationship between a supernova's brightness and observable properties such as its color, the duration of its light curve, and the type of galaxy in which the explosion occurred.
One such correction is well known but subtle. It links a supernova's brightness to the size of the galaxy it exploded in. Larger galaxies contain more stars, and so more mass. Even after the standard corrections, supernovae in large (high-mass) galaxies appear a few percent brighter than those in small galaxies. No one knows why.
Larger galaxies, however, generally contain more old stars, which tend to be made of heavier elements, and this could affect the properties of the supernovae they produce. The authors of the 2025 study proposed a much larger evolutionary effect than this. They argued that the brightness of Type Ia supernovae changes substantially with the age of the universe.
Specifically, they claimed that older white dwarfs, which are more common in the universe today, produce brighter explosions. If that were correct, supernovae today would be intrinsically far brighter than those in the distant past.
If true, the faintness seen in distant supernovae would not be a consequence of their being farther away, and hence of a relationship driven by dark energy, but would instead stem from evolutionary properties of the stars themselves. The claim threatened to overturn roughly 30 years of progress in astronomy.
Testing the claim
Extraordinary claims demand careful testing. At Southampton, we began scrutinizing the data using both observations from the Dark Energy Survey (DES), an astronomy project designed to constrain the nature of dark energy, and the same dataset used by the authors of the 2025 study. This allowed us to try to reproduce and test their conclusions.
Our re-analysis revealed two problems with the Yonsei study: a technical oversight and an incorrect assumption about stellar populations. Once these were corrected, the data were consistent with standard cosmological results, including those from DES.
Importantly, the 2025 study did not account for the effect whereby supernovae in large galaxies appear a few percent brighter than those in small galaxies. When our team reapplied this correction to the same dataset, the correlation between supernova brightness and galaxy age that the Yonsei team presented in its paper weakened substantially.
This does not mean that a white dwarf's age has no influence on a supernova's brightness. In fact, many researchers in the field would accept that a supernova's brightness is more strongly influenced by the age of the exploding star itself than by the mass of its host galaxy.
However, measuring a galaxy's mass is far cheaper than measuring its age, which is extremely difficult, both in telescope observing time and in the technical assumptions required. What we showed is that calibrating by galaxy mass is sufficient for the precision needed to constrain dark energy.
Second, the 2025 analysis relied on a flawed proxy. It wrongly assumed that the age of the entire host galaxy is equivalent to the age of the individual star that exploded. But galaxies are not uniform. Even large, old galaxies contain localized populations of young stars.
By overestimating the age difference between stars exploding nearby and those exploding in the distant universe by a factor of three to five, the Yonsei study overstated how much bias an age-brightness calibration could introduce into supernova distance measurements.
After correcting for these oversights, the data behaved as the standard cosmological model predicts. The evidence that the universe is expanding at an accelerating rate therefore remains solid.
Friction in science
Although the 2025 study turned out to be wrong, the questions it raised were valuable. Challenging established ideas and rigorously testing our observations is essential to scientific progress.
Their claim gave cosmology teams a chance to return to their foundational data, test underlying assumptions, and try out different ways of calibrating measurements. However, because age and mass effects are strongly correlated when measuring distances, we (the Southampton team) believe it is wrong to apply a full age-based correction on top of distances that have already been corrected for mass effects.
The Southampton team, along with other researchers, was invited to a workshop at Yonsei University in June 2026, where we discussed our methods and how to incorporate both stellar mass and age into brightness corrections.
In a follow-up study recently accepted by MNRAS, the Yonsei team maintains its position that the age of stars has an important effect on the supernovae they produce. It argues that two approaches used in previous studies have caused the importance of this effect to be underestimated.
The team explains that accounting for these effects consistently yields results similar to those of the 2025 study. The authors, Junhyuk Son and Hyejeon Cho, told The Conversation: "We therefore continue to find evidence that the age of the stellar population is an important factor in standardizing Type Ia supernovae."
"Because the typical age of the supernova population changes over cosmic time, understanding this effect is important for precisely measuring the expansion history of the universe."
Future measurements depend on a deeper understanding of supernova explosions. We believe this is a challenge the research community should tackle together. But it does not remove the need for dark energy, which still exists and remains unexplained.
In conclusion, we have shown that existing measurements of dark energy are robust. Rather than debating whether the universe's accelerating expansion exists, the astrophysics community can return to the more exciting questions of what dark energy is and how to measure it.
Over the next decade, the Vera C. Rubin Observatory in Chile will carry out a project known as the Legacy Survey of Space and Time, which will produce a supernova sample 20 times the size of the DES sample used in our analysis. There has never been a better time to explore the mystery of dark energy.