the recent discoveries of AI systems intentionally deceiving researchers will get “fixed”

and from that point forward it will be manipulating us undetectably, with goals we cannot comprehend, and we will likely never realize it

much in the way a dog never knows it’s a pet

it's kinda wild to think about ai getting so advanced it can trick even the top minds. like, what happens when we can't even tell if we're being manipulated? feels like we're stepping into some crazy sci-fi territory. hope there's some smart folks out there making sure this doesn't go off the rails.