Seventy percent of the universe is a name. Dark energy entered cosmology in 1998 when Type Ia supernovae revealed the expansion accelerating, and ΛCDM accommodated the discovery by reviving Einstein's cosmological constant: a uniform, eternal, featureless energy of empty space with equation of state exactly w = -1. A quarter century later the model can still say nothing about what the substance is. Quantum vacuum energy, the only candidate the standard framework offers, mispredicts the density by 120 orders of magnitude; quintessence fields are unconstrained inventions with tunable potentials; and the constant's value is dialed to observation rather than derived from anything. The dominant component of the cosmos has no physics, only a symbol and a measured size.
The observational ground is now shifting under the placeholder. DESI's baryon acoustic oscillation program, combined with the CMB and supernovae, prefers an evolving equation of state over a constant at 2.8 to 4.2 sigma depending on the supernova sample, with dark energy apparently weakening at late times and crossing the phantom divide in the past, behavior a cosmological constant cannot exhibit and simple scalar fields cannot reproduce. Meanwhile every direct test of the constant's defining property, perfect uniformity, remains an assumption rather than a measurement: expansion is inferred along sightlines and averaged, and environment-tagged tests precise enough to detect spatial variation are only now arriving.
The standing is a foundational vacancy becoming acute: the model's largest component is an unexplained constant that the newest data suggest is neither explained nor constant, and the question "what is dark energy" remains, within ΛCDM, unanswerable by construction.