If the universe’s history were an ongoing film project in constant post-production, cosmologists would serve as its meticulous editors, endlessly refining the storyline. The current cinematic production is already an awe-inspiring work: it commences with an explosive genesis, where spacetime bursts forth from apparent nothingness. This is followed by the majestic development of stars and, subsequently, galaxies. These celestial structures are shaped by the gravitational force of both visible matter and the enigmatic dark matter, all while the universe peacefully expands, propelled by an unseen entity known as dark energy.
However, this version cannot be the definitive cut. The deeper humanity peers into space, the more the cosmic narrative appears incomplete. The story presents subtle inconsistencies, and key participants remain frustratingly out of reach. For decades, cosmologists have labored to refine this cosmic script.
Now, a fresh spark of inspiration has emerged from the cosmos itself. A powerful new telescope has meticulously mapped millions of distant galaxies, enabling a highly precise tracing of the universe’s expansion. The preliminary revelations suggest that dark energy is behaving in such an unusual manner that it might not be what scientists previously assumed.
This development, if confirmed, represents an exhilarating plot twist. Theoretical physicists are now contemplating a complete reevaluation of dark energy. The ultimate outcome remains uncertain. Nevertheless, many are increasingly receptive to the idea that a richer, more detailed cosmic saga is on the horizon—one that will look notably different from the current rendition.
“We are at a fascinating juncture,” remarks Adam Riess, an astrophysicist at Johns Hopkins University in Maryland, who shared the 2011 Nobel Prize in Physics for his contributions to the discovery of dark energy. He likens the situation to filming a documentary about the creation of our cosmological model, stating, “I would advise: ‘Do not leave your seat now.'”
The Standard Model of Cosmology
Our current comprehensive understanding of the universe’s origins and evolution has been assembled over a century. Its foundation was laid in 1915 with Albert Einstein’s theory of general relativity, which explains gravity as the consequence of massive objects distorting spacetime.
At that time, the prevailing view was that the universe was static. To account for this, Einstein introduced a term he called the “cosmological constant” into his equations. However, in 1929, astronomer Edwin Hubble’s observations of distant galaxies receding from each other indicated that the universe was expanding. This discovery led Einstein to discard his constant.
Following this, the big bang theory emerged. While it is widely accepted today, it was only in the 1960s that the alternative steady-state theory was superseded. This shift occurred when astronomers detected a pervasive field of primordial radiation, the cosmic microwave background (CMB), which perfectly matched predictions for leftover energy from the big bang.
As observational capabilities improved, allowing deeper probes into space, the big bang theory alone proved insufficient. By the 1980s, astronomers observed that the gravitational pull of visible matter was insufficient to hold galaxies together or explain the formation of galactic clusters. The proposed solution was the introduction of invisible dark matter. A decade later, observations of distant supernovae, led by Riess and his collaborators, revealed an unexpected acceleration in the universe’s expansion. The cosmological constant was reincorporated, now termed dark energy.
This framework, essentially, constitutes the current standard model of cosmology, known as Lambda-CDM. The Greek letter lambda represents the cosmological constant, and CDM stands for cold dark matter, hypothesized to be composed of heavy, slow-moving particles. When integrated with general relativity and a few key assumptions—most critically, that the universe appears uniform on large scales in all directions—it provides a compelling explanation for how vast cosmic structures originated from quantum fluctuations in the early universe, amplified by a brief period of rapid expansion known as inflation.
Lambda-CDM is recognized as one of science’s most significant achievements. It harmoniously blends conceptual elegance with an expansive scope, employing just six parameters to delineate the universe’s entire history and generating numerous precise predictions that have been validated by increasingly rigorous observations. “It has been exceptionally successful,” states Mike Turner, a theoretical cosmologist at the University of Chicago. “Comparing it to what we had when I started in cosmology around 1980, it’s far beyond anything we could have imagined. It’s truly astounding.”
Yet, as Turner notes, “it’s now much less than we’re willing to settle for.” This stance reflects the intrinsic nature of scientific inquiry; even the most robust theories are approximations of deeper truths. As they are tested against new data, inconsistencies and areas for improvement inevitably surface.
In the case of Lambda-CDM, the unresolved issues are apparent. Dark matter and dark energy were introduced as preliminary concepts, necessitated by observational evidence but lacking fundamental physical explanations. Despite decades of research, physicists have yet to achieve direct detection of dark matter particles. Dark energy, theorized as vacuum energy arising from quantum fluctuations in empty space, presents significant theoretical challenges. Quantum theory predicts its strength should be approximately 10120 times greater than what is required to drive the observed cosmic expansion.
“Currently, dark energy and dark matter are essentially add-ons,” Turner observes. Both fulfill specific roles, and there is strong empirical evidence for their existence. “But they are purely descriptive, indicating the presence of something more fundamental.”
The Hubble Tension
Fissures have also begun to appear. The most prominent, though with a long history, gained recognition as the Hubble tension in 2015. This discrepancy arises because two distinct methods for measuring the universe’s expansion rate, quantified by the Hubble constant, yield conflicting results. Extrapolating from CMB data using the current model results in a value of approximately 67 kilometers per second per megaparsec. However, direct measurements of the local universe, employing supernovae and variable stars, yield a value around 73. “It represents an end-to-end test of the universe,” says Riess, who posits that the divergence between these measurements strongly suggests fundamental flaws within Lambda-CDM.
Despite this, most cosmologists have been hesitant to abandon the model. All proposed solutions to the Hubble tension thus far tend to diminish the existing model’s nearly perfect alignment with CMB data and the observed large-scale structure. It remains possible that the measurements contributing to this tension contain subtle systematic errors. The methods for measuring late-universe expansion, in particular, rely on a complex chain of inferences, each step dependent on meticulous calibration and assumptions about stars and galaxies. The suspicion is that further data may resolve this tension. “There is too much involved to make a truly definitive statement,” notes Pedro Ferreira, a cosmologist and astrophysicist at the University of Oxford.
Riess disagrees with this assessment. He emphasizes that his late-universe expansion measurements have undergone repeated verification without any identified errors. While some astronomers suggest that independent distance measurements from the James Webb Space Telescope could resolve the tension, Riess counters, “It has been a decade since we discovered the Hubble tension, and it has not disappeared. It has only become more pronounced.”
According to Riess, the primary reason for the scientific community’s reluctance to move beyond Lambda-CDM is the inherent human disposition to retain a theory, especially a successful one, until a superior alternative is available. “People are uncomfortable venturing into the unknown without a clear path.”
Logically, this situation calls for observational evidence that more clearly guides towards a better theoretical framework. Encouragingly, a new generation of telescopes designed to investigate dark energy has begun producing dramatic results, including the Dark Energy Spectroscopic Instrument (DESI).
DESI Results
Installed on a telescope in Arizona, DESI integrates a substantial mirror with 5,000 robotically controlled optical fibers. These fibers rapidly acquire data from distant galaxies, allowing for significantly faster surveys than previous dark energy missions.
Since 2021, DESI has been surveying millions of galaxies to measure their redshift—the degree to which their emitted light has been stretched by cosmic expansion, indicating their distance. By analyzing the characteristic spacing between galaxies at different redshifts, scientists can reconstruct how the universe’s expansion rate has evolved over time.
To calibrate these distances, DESI also measures a subtle imprint from the early universe known as baryonic acoustic oscillations (BAOs). These BAOs, akin to frozen ripples in a pond, preserve a distinct pattern in galaxy separations, providing cosmologists with a “standard ruler” to measure cosmic expansion. The objective was to create the most accurate and precise three-dimensional map of cosmic expansion ever achieved. The latest iteration, released in March 2025 and based on three years of data encompassing 15 million galaxies, has delivered a bombshell, sending tremors through the field of cosmology.
When DESI researchers correlated this data with the latest supernova and CMB measurements, which offer tight constraints on the nearby universe’s expansion, they found that the current Lambda-CDM model does not align as well as a model where dark energy’s strength can vary over time. The primary finding was stark: dark energy appears to be weakening and is not a constant cosmological entity.
“It was quite frightening, actually,” admits Will Percival, an astrophysicist at the University of Waterloo in Canada and a member of the DESI collaboration. He notes that the scrutiny was intense, but adds, “In many ways, this is precisely what people have been anticipating. Experiments that lead us into the unknown and yield unusual, unexpected results are incredibly exciting.”
As if this were not enough, the DESI results also suggest that in the early universe, dark energy might have fallen below the “phantom divide”—a threshold below which its repulsive force would have significantly exceeded that of the cosmological constant—before subsequently increasing again.
“What we are observing with the DESI results, I like to describe as beautifully bizarre,” says Eric Linder, a physicist and cosmologist at the University of California, Berkeley. “Not only do they deviate from the cosmological constant, but they deviate in a manner that was previously unforeseen.” At this stage, the DESI results are not sufficiently robust to claim a definitive discovery. The analysis, at best, favors evolving dark energy with a statistical significance of 4.2 sigma, falling short of the 5-sigma gold standard. This means the finding could still disappear as more data becomes available; the indication of a phantom crossing is even less certain. “I’m on the fence about it,” Ferreira echoes the caution expressed by many in the field, “We’ve just been in this situation so many times before.”
Nevertheless, there are reasons to believe the DESI results might be significant. “It’s the first time I’ve actually exclaimed ‘Ha!'” says Catherine Heymans, an astronomer at the University of Edinburgh. “The method they employ is among the cleanest possible measurements of cosmic expansion we can obtain. It is considerably more difficult to find fault with this than with the Hubble tension.”
This has not deterred critics. In May 2025, George Efstathiou, an astrophysicist at the University of Cambridge, published a paper arguing that the evidence for evolving dark energy is weak for two primary reasons. Firstly, the discrepancy with Lambda-CDM only emerges when supernova data is incorporated into the analysis. Secondly, the DESI team’s statistical analysis relies on predetermined assumptions about the plausibility of different cosmological models, known as “priors,” which Efstathiou contends unfairly favor evolving dark energy models.
However, there is a consensus that if the DESI results gain strength with further data, they would deliver a significant blow to Lambda-CDM. “In that scenario, it’s exciting, because it implies we must re-evaluate our thinking,” says Ferreira.
In a paper published in August 2025, Riess and observational cosmologist Alexie Leauthaud at the University of California, Santa Cruz, argued that the scientific community might be witnessing the decline of Lambda-CDM and must now prepare to move beyond it. Encouragingly, for the first time in 25 years, there is a tangible clue as to what a superior model might look like.
What Will Replace the Standard Model?
This does not imply that deciphering the new paradigm will be straightforward. Although the DESI results have provided a clear direction regarding dark energy’s physical properties, sparking intense theoretical exploration, the resulting depiction of cosmic expansion makes it exceptionally challenging to formulate the correct theoretical description. The most straightforward explanation posits that dark energy originates not from the vacuum but from a field akin to those governing light or nuclear forces. However, these models necessitate suspiciously precise fine-tuning to ensure dark energy intensified in recent epochs rather than at a different time. More critically, they alone cannot account for the phantom crossing phenomenon.
Many theorists favor models where dark energy interacts with gravity, rather than evolving independently. The premise is that gravity’s behavior changes at some point due to energy transfer between ordinary matter and dark energy. “That’s how one can understand that the energy density [of dark energy] could have increased and then decreased,” explains Alessandra Silvestri, a theorist at Leiden University who has demonstrated that such a model aligns better with DESI data than Lambda-CDM. “This is truly the only model that appears to fit.”
Other models propose that dark energy exchanges energy with dark matter, allowing the latter to gradually decay into the former as the universe expands. This concept is particularly appealing from a theoretical standpoint because it links cosmology’s two most significant mysteries.
The challenge with these interacting models is the expectation that evidence for them should have appeared in existing observations, such as those of planetary orbits, which has not been the case. Furthermore, while it is conceivable that the interactions are so minute as to have evaded detection, they might still contravene the fundamental law of energy-momentum conservation.
Consequently, the situation is characterized by a proliferation of ideas, none of which fully resolves the issue. “We genuinely have no definitive answer,” states Ferreira.
For Ferreira and Riess, this suggests that simply attempting to adjust dark energy to better fit the existing data is not the optimal approach. Instead, they advocate for examining what can be learned if the DESI results truly signify the demise of Lambda-CDM. “We should pause and reflect,” suggests Riess. If humanity is indeed at the cusp of a major advancement in understanding the universe, cosmologists must carefully consider how to navigate this transition—not only in terms of long-held assumptions about what constitutes a superior theory but also in how such a theory is discovered.
It is possible that a theory as simple and elegant as the current one will emerge. Alternatively, the most accurate explanation might be more complex, involving a combination of multiple dark energy fields, various types of dark matter, interactions between them, and/or a novel interpretation of gravity on cosmic scales. “This emphasis on elegance and simplicity originates from particle physics,” notes Riess. “But who is to say it applies equally effectively at the scale of the entire cosmos? The universe appears quite intricate from my vantage point, so I believe an open mind is essential.”
Observations will, as always, serve as the ultimate arbiter. DESI continues to gather data, with another release anticipated in 2027. Meanwhile, cosmologists hold high expectations for the European Space Agency’s Euclid space telescope and the Vera Rubin Observatory in Chile, both of which commenced data releases last year. These instruments should provide greater confidence in the developing picture of cosmic expansion, or prompt further revisions. They will also enable observations from previously unexplored redshift ranges.
Ferreira expresses a more tempered outlook. In a 2025 paper, he and his colleagues contended that because cosmological surveys can only examine a limited segment of the universe’s expansion history, numerous theoretical models can produce nearly identical behavior within that interval. As a result, Ferreira believes that even with the influx of new data, “we will be left with a broad family of models that are observationally indistinguishable from the perspective of cosmological data.”
The danger lies in reaching a stalemate akin to the Hubble tension: a situation where many cosmologists hesitate to abandon Lambda-CDM without a more compelling theory, trusting that emerging data might contain overlooked errors, and finding little immediate prospect of developing such a theory. Riess describes this scenario as “Kuhnian purgatory,” referencing philosopher Thomas Kuhn’s concepts of scientific progress, and fears it will lead to stagnation. “Trying to pull the sword from the stone is difficult work… and it might not yield many research papers. But let us not forget that the sword remains lodged in the stone.”
Time for a Paradigm Shift
That being said, Riess suggests the issue lies not with forthcoming data but with the scientific community placing undue emphasis on a pre-existing model rather than the data itself. Whenever a discrepancy arises within Lambda-CDM, its inability to resolve it is often cited as evidence against the new observations. This explains the community’s persistent focus on potential errors and the proliferation of doubt with further measurements. “When you have lived with a standard model for 20 years, many individuals have dedicated the majority of their careers to it,” observes Riess. “Even the notion that this might not be the complete explanation is jarring.”
Perhaps this is simply the nature of paradigm shifts. They are invariably accompanied by conflict, and Lambda-CDM will not yield easily. However, this is not inherently problematic. “You want the defenders to scrutinize any data that appears suspicious. You also want the revolutionaries, those willing to move beyond the current framework,” says Linder. “This interplay, while appearing confrontational, is ultimately healthy.”
Indeed, the very fact that cosmologists are preparing for a potential shift may indicate that a new revolution is truly on the horizon. The one certainty is that after a prolonged period of consensus, cosmology is entering an era of tensions that will render it significantly more compelling. “We are eagerly anticipating all this new data, which I believe will enthrall us all,” states Linder. “It is an incredibly exciting time.”