Two teams measured how fast the universe is expanding. They got different answers, and both are sure they are right. Enter a galaxy's redshift and watch the universe fail to agree with itself — 13.8 billion years old, or 12.7?
Here is a fight that has lasted a decade and nobody is backing down. The universe expands, and the speed of that expansion is the Hubble constant, H0. It is one number. It should have one value. Instead, two elite teams measured it and got answers that refuse to meet in the middle: Planck reads 67.4, SH0ES reads 73.0. That is an 8% disagreement — and 8% is enormous when the two measurements are each supposed to be accurate to under 2%.
Why do they disagree? Because they use completely different rulers. The Planck team measures the baby universe: the afterglow of the Big Bang, whose ripples encode how fast the cosmos was expanding 380,000 years after the start. The SH0ES team measures the adult universe: nearby stars whose brightness is known, then galaxies whose distance is set by those stars, then further galaxies — a "distance ladder" climbed rung by rung out to the present day. One team looks at the past, the other looks at the present, and the present is expanding ~8% faster than the past says it should be.
That gap is the Hubble tension, and it is either a subtle systematic error hiding in one of the measurements, or the biggest clue in modern physics: the universe may be expanding faster now than its ingredients predict. If the tension is real, something is missing from the standard model — maybe new physics in the early universe, maybe dark energy that evolves over time, maybe a weird neutrino. Physicists have published thousands of papers trying to kill the tension or explain it. It survives.
And it is not just an abstract spat. The Hubble constant sets the age of the universe. Planck's H0 says the cosmos is 13.8 billion years old. SH0ES's H0 says 12.7 billion. That is a whole extra billion years of history that exists or does not exist depending on which team you trust. In 2026, the debate has only gotten hotter: new DESI measurements of cosmic expansion and the largest-ever supernova catalogues keep poking at both sides. The universe is still refusing to pick a side.
The universe's expansion rate measured from the early universe (Planck: 67.4 km/s/Mpc) disagrees with the rate measured from nearby galaxies (SH0ES: 73.0 km/s/Mpc). The gap is about 8%, far larger than either team's claimed error — so it is called a "tension" and it will not go away.
Nobody knows. Both teams have checked their work repeatedly, and each new generation of measurements lands on the same two values. If one side has a hidden error, it has survived a decade of scrutiny. If neither has, the standard model of cosmology is missing something.
Possibly — many cosmologists bet on it. But the error would have to be a different one in each method, large enough to close an 8% gap, and invisible across hundreds of independent checks. That is a lot to ask, which is exactly why the tension is still an open problem in 2026.
It changes which age you quote: 13.8 billion years (Planck) or 12.7 billion years (SH0ES). The age is basically 1/H0, so a faster expansion means a younger universe. One team's universe is a billion years older than the other's.