The rock that's in the soil may not be the correct type, or it may have already reacted, or it may be in chunks that are too large to dissolve on human-relevant

The rock that's in the soil may not be the correct type, or it may have already reacted, or it may be in chunks that are too large to dissolve on human-relevant timescales. The idea is that you want ultramafic rocks, which are low in non-reactive Silica but rich in Magnesium and Iron oxides. These oxides absorb CO2 from the atmosphere to form carbonates and bicarbonates, sequestering it for geologic time. However, since reactions can only occur on the surface and when the minerals are dissolved, a big pile of mine spoil or river-sized rocks buried in the ground aren't doing much to absorb CO2. But if we grind it up into rock powder and spread it over fields, or crush it up into sand and dump it on beaches to let the ocean do much of the work of grinding and distributing it, the vastly increased surface area means a significant amount of CO2 can be absorbed in a human lifetime. There can also be pros or cons associated with the minerals in the rocks. For example, you don't want to spread rock containing heavy metals on fields where you grow food. On the other hand, rocks rich in Potassium are a beneficial fertilizer, and Iron-bearing rock spread in the ocean can encourage the growth of phytoplankton that further absorb CO2.