Leap in the dark

Mustapha Ishak-Boushaki recalls the thrill he felt that December night in 2023 when he and two collaborators began unmasking data from the Dark Energy Spectroscopic Instrument (DESI). “We immediately realised we were onto something,” says the Professor of Physics at The University of Texas at Dallas and Co-Chair of the DESI working group that interprets cosmological data. The next morning — on December 13 — as he and his team unveiled their findings to the DESI group, he believed they were close to a ground-breaking revelation. Their data suggested that the rate of expansion of the universe was decreasing, contradicting the standard model of cosmology, which predicts a constant acceleration.

The global collaboration presented its findings based on an analysis of around six million galaxies in 2024. It subsequently expanded the dataset to 14 million galaxies and quasars. As the data doubled, the researchers’ confidence, too, grew. On March 19, 2025, DESI announced its latest results, indicating that the universe might actually be growing weary. The latest result has a confidence level of up to 4.2 sigma, bringing it closer to the 5-sigma threshold, considered the gold standard for confirming a discovery.

Albert Einstein predicted an unknown entity that constantly pushed the boundaries of the universe. Called the cosmological constant (lambda or Ʌ) and later known as dark energy, it makes up 68% of the universe. In 2011, the Nobel Prize in Physics went to three scientists for the discovery of the accelerating expansion of the universe. However, DESI data reveal the weakening nature of dark energy.

“The main takeaway is that dark energy seems (to be) evolving with time and that is putting pressure on our standard model, which is popularly known as lambda (Ʌ) CDM (Cold Dark Matter),” says Shadab Alam, a theoretical physicist at the Tata Institute of Fundamental Research (TIFR), and a collaborator of the DESI experiment. “We might have to come up with a new model,” he adds.

For five years, with the DESI instrument mounted on the Mayall telescope in the Kitt Peak National Observatory in Arizona, the group observed nearly one-third of the sky. Using spectroscopic techniques, the researchers created the largest three-dimensional map of the universe. After the Big Bang, imprinted sound waves known as baryonic acoustic oscillation (BAO) influenced the distribution of galaxies over space. Calculating BAO, the team is studying dark energy by measuring its equation of state, which is the ratio of pressure to energy density. It does this by mapping galaxy distances and their redshifts (speeds), helping scientists track the universe’s expansion history.

The DESI results suggested that dark energy was not a simple cosmological constant— an idea that had long posed contradictions between two major theories: cosmology and quantum field theory. “I would say this is one of the most important findings about the expansion of the universe and its acceleration since the discovery of the acceleration itself in 1998 and will likely have some very important consequences on the future of the standard model of cosmology,” Ishak-Boushaki says.

He adds that existing tensions, such as the Hubble expansion parameter discrepancy, have challenged the standard model. Observations from supernovae, DESI, and the cosmic microwave background (which supports the Big Bang Theory) no longer align as they once did. Allowing for dark energy to evolve over time seems to resolve these inconsistencies better than the static Lambda-CDM model, suggesting cracks in the standard cosmological framework.

As per their calculation, the acceleration rate started diminishing a few billion years ago. Following this trend, the expansion may stop accelerating, and may come to a halt. “If we extrapolate and calculate when the universe will stop accelerating and dark energy will no longer be dominant, we can determine it (the time scale) easily,” he adds.

Standard cosmology assumes that the universe is homogeneous and isotropic. However, researchers have shown that the universe is not uniformly smooth everywhere (bit.ly/anisotropy-space). Subir Sarkar, a cosmologist at the University of Oxford, notes that the inference of dark energy is based on assuming isotropy. DESI analyses should first check if this is true, given the recent high-significance evidence for anisotropy (the opposite of isotropy) in the distribution of quasars and radio sources. “The headline conclusion that dark energy is evolving follows from including supernova data, but here too there is recent evidence for anisotropy,” Sarkar says.

See also:

Chipping away at dark energy