Why switch to non-fissile thorium when uranium is cheap and abundant, now and for at least the next fifty years and we already have a lot of experience using it to fuel existing reactors and their replacements currently under construction? Of course at the end of that time we can reprocess the spent fuel that's in store and fuel another generation of uranium reactors for fifty years and more from the "waste". If all else fails there are proven techniques to extract uranium from seawater, cost estimated to be about USD 300 per kilogram.
Thorium is difficult to use as a nuclear fuel. It's not fissile, it needs to be converted in a high-temperature high-neutron-flux breeder cycle into U-233 to make it into a fuel that can produce energy by being fissioned. The theory, and it is only a theory, is that this can be done in a continuous process. Further theory suggest it can be done in a molten-salt stream but it's never been achieved in the real world, just in Powerpoint slides and at TED talks.
No, the molten-salt test reactors that were run at ORNL and elsewhere fifty years ago didn't use thorium, they used U-233 and later U-235. Some reactors have used thorium, usually pebble-bed designs which were not a great success (see the German HTHR-300 for an example) and they were mostly fuelled by U-235 and in some case Pu-239/240 with a little added thorium to season for taste.