If it's not practical to mitigate against some risk, then the potential reward isn't high enough. Ignoring the risk and continuing without optimal mitigation is

If it's not practical to mitigate against some risk, then the potential reward isn't high enough. Ignoring the risk and continuing without optimal mitigation is only a logical course of action if the potential reward outweighs the possible downside. If research is sufficiently dangerous to necessitate a BSL-3+ lab, then it should have to meet a correspondingly high bar for demonstrating its potential reward. It's not clear to me that gain of function research could ever make such a justification. In an ideal world, we could do this research with remote controlled robots in an orbiting space capsule. If we can't do that yet, then we could at least do it in a desert. Otherwise, shouldn't we be asking the local population to approve each research proposal that could lead to a local outbreak of a dangerous virus?