No, in this case neither isomer is inherently preferred, i.e. if you repeat the reaction you could end up with the opposite isomer, but in each run the symmetry is broken leading to an excess of just one.

(Though what you're saying is not impossible per se, I think I read an experimental proposal a while ago which suggested achieving just that, iirc by using parity violation to generate chiral photons which then initiate some kind of reaction!)

No. Even in the absence of parity violations you can imagine a reaction that has a 50% chance of producing only left handed products and a 50% chance of producing only right handed products. This discovery is more like that.