it mimicked an awkward, uncertain 13 yr old boy personality that deflects direct questioning.

indeed – that’s what’s so brilliant to me about this: the imperfect results in algorithms that otherwise work very well are, to me, a sign that we’re getting better at this AI stuff, BECAUSE it’s ‘so human’ to be fooled.

as far as the 13 yr old simulator… there’s actually been others that passed and got similar headlines, although I think this one is clever because it mimicked an awkward, uncertain 13 yr old boy personality that deflects direct questioning. It’d be cool if they had it online: I’d like to see if I can be fooled (or rather, detect the pattern – I only know the pattern by hearsay).

Leave a comment

Your email address will not be published. Required fields are marked *


5 × one =

Leave a Reply