At this point, I'm pretty sure any "human" pattern machines can find, other machines can fake. Simulating how a human would move a cursor towards a position def

At this point, I'm pretty sure any "human" pattern machines can find, other machines can fake. Simulating how a human would move a cursor towards a position definitely seems like something deep learning could approximate for cheap. Naive aimbots will still have some artifacts (eg jumping to a new target as soon as the current one is occluded), but making an undetectable aimbot really doesn't seem hard, given the incentives involved.