pebkac thoughts

I’ve been thinking a bit about Paul’s post on AI surpassing human intelligence, and I realized that the most interesting test of whether or not AI has come of age would be whether or not it could lie. Passing human intelligence, no matter what type of intelligence, doesn’t bring with it the sort of interesting problems that lying does. If you think about it, lying implies agency, context, and desire. Has somebody already written about this? Obviously this has appeared in countless SF works, right? …

Off to google I go.

UPDATE: Looks like Turing thought of that first, too (last paragraph).