I doubt it. At any rate, I wasn't claiming to offer "the strongest argument that AI existential risk is a big problem". I wasn't claiming to offer any argument

I doubt it. At any rate, I wasn't claiming to offer "the strongest argument that AI existential risk is a big problem". I wasn't claiming to offer any argument that AI existential risk is a big problem. I was pointing out an interesting feature of the argument in the comment I was replying to: that (so far as I can see) its reason for dismissing AI x-risk concerns would apply unchanged even in situations where AI x-risk is in fact something worth worrying about. (Whether or not it is worth worrying about here in the real world.)