Dark energy is the name physicists give to whatever is pushing the universe's expansion to speed up. It makes up roughly two-thirds to 70 percent of everything in the cosmos — more than all the stars, planets, gas, and dark matter combined — and it remains the biggest unsolved mystery in physics. The leading idea is that it is simply a property of empty space itself, and the newest maps of the universe, released in 2025, suggest it may even be weakening over time.
The discovery nobody expected
The story begins in 1998 with two rival teams hunting exploding stars. The Supernova Cosmology Project, led by Saul Perlmutter, and the High-Z Supernova Search Team, led by Brian Schmidt and Adam Riess, were measuring Type Ia supernovae — exploding stars so consistent that they work as "standard candles" for judging cosmic distances. Both teams expected the same thing: because gravity pulls everything together, the universe's expansion should be slowing down.
The data said the opposite. More than 50 distant supernovae were dimmer — therefore farther — than they should have been in a decelerating universe. The expansion was not slowing down. It was speeding up. Something invisible was pushing galaxies apart faster and faster, against gravity's pull. The teams published within weeks of each other, and the 2011 Nobel Prize in Physics went to Perlmutter, Schmidt, and Riess for overturning the old picture: instead of a Big Crunch, the universe's likely fate became a long, cold, accelerating expansion.
A constant — and the worst prediction in physics
The simplest explanation for the acceleration was already on the shelf. Einstein had once added a "cosmological constant" to his equations to represent an intrinsic energy of empty space, then famously discarded it as a blunder. Dark energy revived it: if space itself carries energy, and that energy has negative pressure, then as the universe expands and creates more space, the total push grows stronger — a runaway expansion.
This is where physics hits a wall. Quantum field theory predicts that the vacuum should be full of zero-point energy — virtual particles flickering in and out of existence in every mode of every field. When theorists add it all up, the predicted vacuum energy density overshoots the observed value of dark energy by roughly 120 orders of magnitude. That is not a small disagreement; it is routinely called the worst theoretical prediction in the history of physics. A naive calculation of how much the vacuum should weigh gives an answer larger than observation by a factor of one followed by 120 zeros. No one knows how to fix it — whether some unknown mechanism cancels almost all of it, or whether the naive estimate simply does not mean what it seems to.
The observed universe, meanwhile, behaves as if dark energy is constant. That baseline model has a name — Lambda-CDM — and its parameters have been measured with exquisite precision by the cosmic microwave background and galaxy surveys. For over two decades, Lambda-CDM kept winning every test.
DESI's 2025 hint: what if it is changing?
In March 2025, the Dark Energy Spectroscopic Instrument (DESI) collaboration released its second major data set — and cosmology felt its first real wobble. DESI, mounted on a telescope in Arizona with 5,000 robotic positioners measuring thousands of galaxies at once, has built the largest three-dimensional map of the universe ever made: about 14 million galaxies and quasars stretching back 11 billion years. It tracks dark energy by measuring baryon acoustic oscillations — frozen ripples from the early universe that act as a standard ruler, letting astronomers measure the expansion rate at different epochs.
Combined with microwave-background and supernova data, the DESI measurements hint that dark energy may be evolving — weakening over cosmic time rather than staying perfectly constant. In the standard parametrization of the dark energy equation of state, the data drift away from the constant-energy point with roughly three-sigma significance, depending on which supernova sample is used. That is enough to make the community sit up, but it is not a discovery: particle physicists require five sigma for that. It also depends on supernova measurements that are themselves under scrutiny, with debates about calibration and whether the ages of host galaxies subtly bias the results.
So the honest scorecard: the cosmological constant remains the favored baseline, but the hints of something dynamical are the strongest challenge the standard model has faced in 25 years. The difference matters enormously. If dark energy is constant, the universe expands forever, faster and faster. If it is evolving — a dynamical field sometimes called quintessence — the future could hold surprises nobody has modeled yet.
What could it be?
The candidates fall into three camps. The first says dark energy is genuinely a cosmological constant, and the DESI hints will melt away as systematics are understood. The second says it is a dynamical field — a new form of energy whose strength changes with time, perhaps slowly rolling downhill like a ball on a slope. The third says there is no new energy at all, and our theory of gravity breaks down on cosmic scales; modified-gravity theories try to reproduce the acceleration without adding anything to the energy budget.
Each camp has a problem. The constant has the 120-orders-of-magnitude embarrassment. Dynamical fields fit the DESI data better but add complexity that the data does not yet demand. Modified gravity must also survive the ultra-precise tests that general relativity passes in the solar system. Meanwhile, the same surveys are sharpening other tensions — the disagreement over the Hubble constant, and the impossibly massive early galaxies JWST keeps finding — which may or may not be connected.
Here is what dark energy rules and what it does not. It rules the universe's present and future: roughly 70 percent of the energy budget, the deciding vote on cosmic destiny. It does not rule our daily lives — its effect is utterly negligible on anything smaller than intergalactic scales. It is the dominant ingredient of reality and the least understood, a pressure we can only see in the motion of everything, everywhere.
The next few years will be decisive. DESI is only partway through its planned 40-million-object survey, and next-generation maps will shrink the uncertainties. If the weakening signal survives, physics gains a new field to explain — and the "worst prediction in history" may turn out to have been pointing at the answer all along.
