The universe is expected to be homogeneous on a large scale, but DESI data suggest otherwise

Introduction: New Data from the DESI Instrument

Earlier this year, the Dark Energy Spectroscopic Instrument (DESI) completed major observations. These observations mapped 47 million galaxies over a staggering distance of 11 billion light-years. This technical feat offers astronomers the opportunity to better assess the large-scale structure of the visible Universe.

It was while studying this colossal dataset that astronomers Francesco Sylos Labini and Marco Galoppo made an unexpected discovery. According to a study published in the scientific journal Nature in 2026, their work suggests that the universe may not look the same in all directions. This observation directly contradicts one of the fundamental assumptions of modern cosmology.

This new report calls on the scientific community to rethink how matter is distributed throughout the cosmos. The results obtained by this team raise profound questions about the very structure of our cosmic environment, far beyond what previous instruments had been able to measure until now.

The Cosmological Principle: A Fundamental Rule

On the scale of a single galaxy or local groups of galaxies, the universe clearly appears to be anisotropic. This scientific term means that the structure differs depending on the direction in which the observer is looking. For example, in one specific direction, space may appear mostly empty, while another direction may contain a dense galaxy cluster.

However, the cosmological principle posits that on larger scales, the Universe consists of matter distributed more or less uniformly in all directions. This concept is directly based on the Copernican principle, which states that there should be no special observer in the Universe. In other words, on a large scale, the cosmos should look the same from any vantage point.

To put this concept in layman’s terms, imagine the universe as a piece of fabric. When zooming in to the scale of individual fibers, you can see empty spaces and interwoven threads forming an overall structure. However, when you zoom out to observe this fabric on a much larger scale, it appears uniform, and its matter seems to be distributed homogeneously throughout.

The Scientific Debate on Isotropy

The exact scale at which the universe is expected to appear isotropic has long been the subject of heated debate within the scientific community. Galactic surveys have revealed a “cosmic web” composed of filaments, walls, and voids. Given these observations, scientists are unsure exactly how quickly this structure is supposed to fade as the scale of observation increases.

Some research, particularly studies focusing on the cosmic microwave background, has provided evidence supporting the cosmological principle. At the same time, other studies have shown that an anisotropic structure does indeed persist on scales ranging from tens to hundreds of megaparsecs.

However, the statistical significance of these conflicting studies has remained uncertain until now. This uncertainty has prompted astronomers to seek new analytical methods capable of resolving the issue with greater precision, drawing on massive datasets.

A novel method revealing an unexpected structure

The authors of this new study point out that previous surveys of anisotropy focused primarily on preferred directions, rather than assessing a more general directional structure. To conduct more comprehensive tests and measure how the distribution of matter varies with distance and angle, they used a statistic called the angular distribution of pair distances (ADPD).

This is a parameter-free statistic that allows for rigorous measurement of directional correlations. By comparing their results with a theoretical model based on expected isotropy, the researchers made a major discovery. They found that the galaxy samples from DESI show a persistent anisotropic structure in the distribution of galaxies, even on scales of the order of a gigaparsec.

In other words, galaxies cluster more than they should at scales far larger than those examined previously. Building on earlier studies that suggested anisotropy on the megaparsec scale, this new study indicates that anisotropy persists at scales 1,000 times larger. “These results provide direct evidence that directional coherence persists on scales larger than those predicted by the standard model, challenging the assumption of large-scale isotropy,” write the study’s authors.

Toward a Revision of Foundational Theoretical Models

Although these results are impressive, the researchers acknowledge certain limitations to their study, which on its own cannot determine the physical origin of this anisotropy. It remains possible, moreover, that if these data are confirmed, the Universe may eventually become isotropic on even larger scales. Nevertheless, since the cosmological principle underlies many foundational concepts, the authors warn that several ideas will need to be revised if this larger-scale anisotropy is definitively confirmed.

In their report, they explain in detail: “As such, this detection of large-scale anisotropies contrasts with the standard formulation of the cosmological principle, which assumes statistical homogeneity and isotropy around any point, while remaining compatible with the Copernican principle, which requires only the absence of privileged observation points.”

The scientists conclude by discussing the future implications of their work: “Next, from a theoretical standpoint, the existence of such large-scale anisotropies motivates the exploration of more general solutions to Einstein’s field equations that explicitly allow for large-scale inhomogeneities as cosmological models and/or the investigation of alternative sources of accelerated structure formation, for example, through the introduction of self-interactions in the dark matter component or feedback effects arising from inhomogeneities.” This groundbreaking work, led by Francesco Sylos Labini, is documented under the identifier 10.1038/s41586-026-10702-5.

According to the source: phys.org

The universe is expected to be homogeneous on a large scale, but DESI data suggest otherwise