Stuart_Armstrong comments on AI risk, executive summary

[Stuart_Armstrong]

Thanks! I’ll think about your general point; the specific example seems weak, though.