Author here. Unfair point, article makes clear I don't expect "outdated verification methodology" to catch up - and I even link that pessimistic expectation wit

Author here. Unfair point, article makes clear I don't expect "outdated verification methodology" to catch up - and I even link that pessimistic expectation with the issue of emergence that fatally undermines NNs. Point about lab vs reality is fair.