Bonus points if your static site sends a 503 with a retry after header.
Maybe not the hardest, but still challenging. Unknown biases in training data are a challenge in any experimental design. Opaque ML frequently makes them more challenging to discover.
Depends heavily on the kind (and intensity) of radiation. Beta (electron/positron) and gamma (photon) generally won’t, but neutron and alpha can. Many of the atoms that become radioactive will rapidly decay, and that’s one of the mechanisms behind the impact to structural integrity.