A critical look at catastrophe risk assessments

Recent papers by Busza et al. (BJSW) and Dar et al. (DDH) argue that astrophysical data can be used to establish bounds on the risk of a catastrophe in forthcoming collider experiments. The safety case set out by BJSW does not rely on these bounds, but on theoretical arguments, which BJSW find sufficiently compelling. However, DDH and other commentators (initially including BJSW) have suggested that the astrophysical bounds alone do give sufficient reassurance. This seems unsupportable when the bounds are expressed in terms of expected cost. For example, DDH's main bound, $p_{\rm catastrophe} < 2 \times 10^{-8}$, implies only that the expectation value of the number of deaths is bounded by 120. We thus reappraise the DDH and BJSW risk bounds by comparing risk policy in other areas. We find that requiring a catastrophe risk of no higher than 10^{-15} is necessary to be consistent with established policy for risk optimisation from radiation hazards, even if highly risk tolerant assumptions are made. A respectable case can be made for requiring a bound many orders of magnitude smaller. We conclude that the costs of small risks of catastrophe have generally been significantly underestimated in the physics literature. This conclusion does not affect the theoretical arguments against the possibility of catastrophe in the experiments considered. It does, however, suggest more careful consideration of the degree of confidence that can reasonably be placed in those arguments.