For more than two decades, one idea has anchored our picture of the cosmos: the universe isn’t just expanding, it’s speeding up, driven by a mysterious “dark energy” baked into empty space itself. That conclusion won the 2011 Nobel Prize in Physics. Now a new analysis is asking whether it was right all along.

A team led by researchers at the Tata Institute of Fundamental Research in Mumbai, working with Professor Subir Sarkar of the University of Oxford, revisited Pantheon+, one of cosmology’s most important datasets, containing observations of more than 1,700 exploding stars known as Type Ia supernovae. These stellar explosions have long served as cosmic “standard candles,” letting astronomers measure how the universe’s expansion has changed over billions of years.

The twist: the team applied a recently proposed correction that accounts for the age of the stars that produce these supernovae. Once that correction was factored in, the signal for acceleration didn’t just weaken, it flipped. The data pointed toward a universe that is decelerating, not speeding up.

The researchers also tested whether the apparent acceleration looks the same in every direction, an assumption baked into the standard cosmological model. It didn’t. They found the effect was strongest in the direction our local region of space is already moving (as tracked by a hotspot in the cosmic microwave background), and faded with distance. Sarkar argues that’s a problem for dark energy specifically: an effect from the quantum vacuum shouldn’t care which way you’re looking.

Not everyone agrees. In the very same journal issue, a separate team including Oxford’s Professor Maria Vincenzi published a rebuttal defending the standard picture, arguing the evidence for acceleration remains solid and that the debate should now shift to understanding dark energy’s true nature, not questioning its existence.

So who’s right? Neither side has the last word yet. The real test may come from the Rubin Observatory’s Legacy Survey of Space and Time, which is expected to catalogue hundreds of thousands of supernovae in the coming years, a dataset large enough to settle whether the universe’s acceleration is real, directional, or an illusion born from how we measure exploding stars.

Sources