When it comes to geological periods of time, our intuitions about how things work cannot be trusted. This is a reflection of the parochial character of many of the heuristic shortcuts in our minds. The same thing applies to the behaviour of objects at a minute scale. For instance, sufficiently tiny machinery is hampered enormously more by friction and surface tension than a larger equivalent would be. Because they have more surface area relative to their volume, they also tend to be much more reactive.
Indeed, asymmetries of behaviour at different scale raise serious concerns about the safety of newly developed nanotechnologies. Just as our brains are calibrated to deal with the kind of experiences that have been normal to human lives for thousands of years, our regulatory procedures are calibrated to respond to known risks like toxicity or corrosiveness.
There have certainly been serious problems that arose from regulation lagging innovation in the past. Think of ozone-destroying chlorofluorocarbons, or mesotheliomas caused by chrysotile asbestos. Balancing safety concerns with the desire not to stifle innovation is extremely challenging, especially when the entities with the most sophistication in relation to a new technology are its commercial backers.
In some cases, nanomaterials have almost completely escaped regulation because it has been assumed they behave like their non-nanoscale equivalents. That said, nanoscale titanium dioxide is not the same as a macroscopic bar of the stuff. The same is true for carbon nanotubes, silver nanoparticles, and so forth. Indeed, if the substances were equivalent, there would be no promise in nanotechnology itself. Especially when it comes to the exposure of nanoparticles to human beings (though food, cosmetics, etc), it makes sense for the nano-versions to be regulated as new substances, with the onus on the manufacturers to demonstrate safety.