The AI bio risk conversation would be less contentious if we talked less about x-risk and more about 'mere' catastrophic risks. Human extinction is just incredibly difficult. Positing bio attacks that kill thousands or millions is more plausible and might generate less pushback.