NASA's Roman Telescope May Overturn Our Understanding of Dark Energy

In the late 1990s, astronomers made a discovery that rewrote the textbooks. Using data from the newly launched Hubble Space Telescope and observations from ground-based instruments, they found that the universe was not slowing its expansion as physics had long predicted. It was accelerating. Something invisible was pushing galaxies apart at an ever-faster rate. Scientists called it dark energy, and the researchers who documented this phenomenon—including Australian astrophysicist Brian Schmidt—won the Nobel Prize in 2011 for their work.

But nearly three decades later, that hard-won certainty is beginning to crack. New measurements from recent experiments, including the Dark Energy Spectroscopic Instrument and the Dark Energy Survey, are hinting at something troubling: dark energy may not be constant at all. Instead of maintaining the same strength from the Big Bang to today, it might actually be weakening over time. If true, this would overturn not just our understanding of dark energy itself, but the entire mathematical framework physicists use to model the universe—a model called Lambda-CDM that has anchored cosmology for two decades.

NASA's newly launched Nancy Grace Roman Space Telescope may finally settle the question. The telescope, which began operations in late August 2026, carries an infrared camera designed to capture vastly larger patches of sky than its predecessors, allowing it to spot distant galaxies and rare stellar explosions in numbers never before possible. Within three months of launch, Roman will begin sending back its first images and conducting science operations. Within five years, it should have enough data to either confirm the unsettling hints from DESI and other experiments, or rule them out entirely.

Cullan Howlett, an astrophysicist at the University of Queensland, explains why this matters so profoundly. Galaxies are not scattered randomly across space. They cluster together, connected by filaments of matter across vast voids—a structure sometimes called the cosmic web. This three-dimensional pattern encodes crucial information: the ingredients needed to build a universe, and how fast that universe has expanded since its beginning. The more detailed the map of this web, the easier it becomes to measure how dark energy and its equally mysterious counterpart, dark matter, have shaped cosmic history. Neither phenomenon interacts with ordinary matter in ways that make them directly observable, yet physicists know both exist because they can see their gravitational effects on how galaxies move through space and time.

The stakes of getting this right are almost unimaginably high. If dark energy is indeed weakening, the universe's expansion will eventually slow down. If, conversely, dark energy is accelerating faster than current models suggest, the universe could eventually tear itself apart in what physicists call the Big Rip. The choice between these futures depends entirely on what dark energy actually is—and right now, nobody knows. Ami Choi, a NASA astrophysicist, compares the potential shift to the upheaval that occurred when Einstein's theory of general relativity replaced Newton's laws of motion. Scientists call such fundamental overturnings "paradigm shifts." One may be coming.

Yet caution is warranted. Cullan Howlett acknowledges that the recent hints of weakening dark energy could be artifacts of measurement—quirks in how DESI's instruments work, or biases in which portions of the sky the experiments studied. Hundreds of astrophysicists have worked on these projects with rigor and care, but independent verification remains essential. "If you detect something really exciting with one telescope or one experiment, you want to be able to confirm that with an independent dataset," Howlett says. That is precisely what Roman is designed to do.

Roman's wide-angle camera will help researchers spot a particular type of supernova—stellar explosions that occur in a consistent, predictable way and can therefore be used to calibrate cosmic distances. These events are rare and fleeting, making them difficult to find. But Roman's ability to photograph enormous swaths of sky in a single exposure dramatically increases the odds of catching them. Brad Tucker, an astrophysicist at the Australian National University, notes that for supernovas and dark energy research, "it's a numbers game." Roman will also enable researchers to use gravitational lensing—the bending of light by massive objects—to study the universe in ways ground-based telescopes cannot match.

NASA administrator Nicky Fox has called Roman "a discovery machine that will bring us closer than ever before to answering humanity's most profound questions about our cosmic history." Whether the telescope confirms the unsettling new data or refutes it, Choi says, the outcome will advance understanding. "Anytime you get a different answer from another experiment, you learn something," she explains. Either way, the next chapter of cosmology is about to be written.