Our universe has been expanding ever since it was born in the Big Bang. For decades, most cosmologists believed the universe’s expansion was slowing. Gravity, pulling on all matter, would eventually overcome the momentum from the Big Bang, halting or even reversing the expansion.
But in 1998, supernova observations revealed a surprise: the universe is not just expanding; it is accelerating. To explain this phenomenon, scientists proposed a mysterious repulsive force pushing galaxies apart, called dark energy. For the next 25 years, most assumed this force was constant, an inherent and unchanging property of space driving it to expand ever faster. This force is now part of our standard model of cosmology.
Now, recent findings released March 19 from two of the largest cosmological surveys to date — the Dark Energy Spectroscopic Instrument (DESI) survey and the Dark Energy Survey (DES) — challenge that long-held assumption, adding to a mounting body of evidence that suggests the force from dark energy may not be constant after all. Instead, the density of dark energy may vary over cosmic time.
Related: Supernova survey hints dark energy could be changing
Hints of deviation
“When the experiment was starting, the assumption was that we would just obtain the highest-precision measurement confirming the constant,” says Stephanie Juneau, an astronomer at NSF’s NOIRLab and a DESI data team member. “It was actually a huge surprise that we found some hint of deviation.”
The evidence emerges not from any single dataset, but from a growing convergence across independent measurements. DESI researchers analyzed a three-dimensional map containing almost 15 million galaxies and quasars — the most detailed spectroscopic map of the universe ever created. On its own, DESI’s data do not significantly challenge the standard cosmological model. But when combined with external measurements, including data from the cosmic microwave background (the afterglow of the Big Bang), supernovae studies, and gravitational lensing, tensions begin to emerge.
Meanwhile, DES scientists, using separate methods across six years of observations, found similar anomalies that suggest the standard model may not tell the full story.
“What was most convincing for me, is that the evidence is coming from different directions. And it can’t be that all these different directions of different data sets conspire to give the same wrong answer,” says Mustapha Ishak-Boushaki, a theoretical astrophysicist at the University of Texas at Dallas and co-chair of the working group that analyzed the DESI data.
Related: Are the percentages of dark matter and dark energy stable?
Changing, not constant
The idea that dark energy might be changing over time is more than a provocative twist; it addresses one of the deepest puzzles in modern physics. In Einstein’s equations describing the universe, the cosmological constant Lambda (Λ) — now believed to represent dark energy’s force — offers a straightforward way to explain accelerated expansion. Quantum field theory predicts that empty space should nonetheless contain energy, called vacuum energy. This is one candidate for dark energy, but theory predicts that vacuum energy should act like a repulsive force more than 40 orders of magnitude stronger than what astronomers observe.
“If [dark energy] is a cosmological constant, it has been considered a dead end,” Ishak-Boushaki says. “The difference between theoretical calculations and measurements was known in the field as the most embarrassing difference that we can’t explain.”
Evolving dark energy, though, could potentially resolve this contradiction between dark energy’s tiny measured value and the vacuum energy that quantum theory predicts. If dark energy changes over time, that gives astronomers entirely new theoretical frameworks to explore, from modified theories of gravity to energy fields that naturally evolve throughout cosmic history.
“With DESI, we saw the light,” Ishak-Boushaki says. “The hope is back that it is not a cosmological constant, it’s something different. Do we know what it is exactly now? No, but the door has opened for us to find it.”
Given that dark energy makes up about 70 percent of the universe, confirming that it changes over time would mark a profound shift in our understanding of the cosmos’ fate. The cosmological constant model implies a universe that expands ever faster, eventually becoming cold and empty in a scenario sometimes dubbed heat death or the Big Freeze. But dynamic dark energy could lead to radically different outcomes, slowing expansion or accelerating it so violently that everything ends in a catastrophic Big Rip, tearing apart galaxies, stars, and even atoms.
The current level of statistical confidence in the combined results — around 4.2 sigma — is approaching the 5-sigma threshold typically required for a scientific discovery to be accepted. At that level, the probability of the result being a statistical fluke drops to less than one in a million.
“For people like me working on this for 25 years, we didn’t expect this to happen in our lifetime,” Ishak-Boushaki says.
More data needed
Still, the researchers caution that such predictions remain speculative. Much depends on whether dark energy really is changing and, if so, what’s driving that evolution.
“I’m not going to hold a funeral for the standard cosmological model,” says Jessie Muir, a cosmologist at the University of Cincinnati and member of both the DES and DESI teams. “I’m cautiously excited. It’s something to keep an eye on and prod at and really try to make sure we understand.”
The next few years will be critical for exploring the tentative findings. DESI plans to continue collecting data through 2026, aiming to map over 50 million galaxies and quasars and potentially push the evolving dark energy signal beyond the 5-sigma discovery threshold. Researchers also plan to cross-check findings with new instruments and experiments, including upcoming projects like the Vera C. Rubin Observatory’s Legacy Survey of Space and Time and NASA’s Nancy Grace Roman Space Telescope. These tools will allow scientists to probe deeper into space and time than ever before, presenting unprecedented opportunities to investigate the nature of dark energy.
“It gives us a chance to go back to the drawing board. And then we get to learn, really, what’s the true nature of a universe,” Juneau says. “I see an immense reservoir of new discoveries that are waiting to be made.”
