The water leaving the mix
Fresh concrete carries more water in it than the cement needs to cure. Over the first weeks, that extra water works its way out and the slab shrinks very slightly as it goes, a fraction of an inch over the length of a driveway, but enough that something has to give.
This is not a defect in the mix or a sign of a bad pour. Every slab of concrete does this to some degree as it cures. What differs from job to job is where that crack ends up. A slab poured in hot, dry, windy weather loses that surface moisture faster than one poured on a mild day, and losing it too fast is part of what makes that early cracking more likely.
What the slab is holding onto while it shrinks
Shrinkage on its own would be harmless if a slab could pull inward evenly from every side. It cannot. The underside grips the gravel and soil it was poured over, and the edges sit against a foundation wall, a garage step, a footing, or an older slab that is not going anywhere. Everything the concrete touches holds it where it is while the middle tries to pull away, and that restraint is what turns a small amount of shrinkage into a crack.
Anything cast into the slab does the same thing at a smaller scale. A pipe sleeve, a floor drain, a post set through the concrete, or a section that thickens where it passes over a footing gives the slab a spot that is stiffer or thinner than the concrete around it, and the pull concentrates there. Inside corners are the plainest example. An L shaped patio, or a slab notched around a step, very often cracks straight out from that inside corner, because that is the one place the tension has nowhere to spread.
Because a shrinkage crack follows the weakest line available to it, a control joint is cut early to give it a straight one. How that grid is planned, and why the timing of the cut matters, is a question of its own with its own answer.
The ground moving every season after
Curing shrinkage is only the first half of the story, and it is over within the first weeks. Once the slab has cured, the subgrade underneath it keeps moving with the seasons, swelling with moisture and settling as it dries, freezing and shifting through winter at this elevation. That movement puts a slower, ongoing kind of stress on the slab that has nothing to do with the original pour.
A slab sitting on firm, well compacted ground barely notices this. One sitting on a subgrade that was never properly prepared feels every one of those seasonal shifts, which is why two driveways poured the same week out of the same truck can age very differently. This is also the part nobody sees once the concrete is down, and the part that is nearly impossible to fix afterward without taking the slab back up.
Reading a crack back to which of the three caused it
Where a crack sits usually tells you which of the three it came from. A thin line sitting inside one of the cut joints is shrinkage, released exactly where the joint was cut to release it. A crack running out from an inside corner, from a post, or from the edge of a drain is restraint, the slab pulling against something that would not let it move. A crack that wanders across open concrete, showing up years rather than weeks after the pour, is usually the ground.
None of the three means the concrete was poured badly, and none of them is unusual. What they change is what happens next. Shrinkage at a joint gets left alone. A corner crack is worth watching to see whether it stays tight. Movement in the ground is the one that earns a real look, because the crack on the surface is the symptom and the reason for it is underneath.
Where the line to something else sits
A crack that sits higher on one side than the other, or keeps widening season after season, is telling a different story than any of the three above. That points at the ground moving in a way the slab was not built to take, which is a separate question from the ordinary cracking every slab does, and it does not settle down on its own the way curing shrinkage does.