Utilizing AI / Machine Learning and Datasets for the Future of Aimbot?

[Injected]

You have to realize more humanization you add. More shitty the Aimbot will get. It will have to miss > Overshoot > Over Correct LIKE A HUMAN. None is liked by anyone who cheats these days.

Not to mention there is 0 guarantee what we pre-create will be able to bypass AI Anti Cheat from the get-go. It could be instantly detected rendering all work done useless.

Unless we have something to test against it's not worth our limited time.

[Wallhacks]

You have to realize more humanization you add. More shitty the Aimbot will get. It will have to miss > Overshoot > Over Correct LIKE A HUMAN. None is liked by anyone who cheats these days.

Not to mention there is 0 guarantee what we pre-create will be able to bypass AI Anti Cheat from the get-go. It could be instantly detected rendering all work done useless.

Unless we have something to test against it's not worth our limited time.

That certainly could all be true, but the part about the aimbot aiming and shooting completely for you and looking legit is the best part. With that being said, time is limited since you do support a vast amount of games and this certainly wouldn't be a priority, although it'd set you apart from all (afaik) other providers. Also, aimbot being humanized doesn't mean that you have to miss so many shots that you lose a fight. It just means every here and there the aimbot might miss when you have settings set to say "Top 1% player" in a drop down menu. For now, we all just have to use very obvious rage settings when in a sticky situation that doesn't look like top 1% player tracking for example. No one that I know of has this while looking legit. For instance, if you activate a rage mode, it will snap and hold onto the target whereas what I think would be neat is that if an enemy player is wrapping around you, it will accelerate towards them and start firing and may move the mouse slightly past them or to the edge of their body in the direction they're wrapping around you and then settle as a real player would (instead of locking onto the body that looks obvious)...but all of this with no user mouse movement required. Completely on autopilot for the situation one wants to activate the mode while looking real at the same time. This would be a gamechanger.

[Injected]

That certainly could all be true, but the part about the aimbot aiming and shooting completely for you and looking legit is the best part. With that being said, time is limited since you do support a vast amount of games and this certainly wouldn't be a priority, although it'd set you apart from all (afaik) other providers. Also, aimbot being humanized doesn't mean that you have to miss so many shots that you lose a fight. It just means every here and there the aimbot might miss when you have settings set to say "Top 1% player" in a drop down menu. For now, we all just have to use very obvious rage settings when in a sticky situation that doesn't look like top 1% player tracking for example. No one that I know of has this while looking legit. For instance, if you activate a rage mode, it will snap and hold onto the target whereas what I think would be neat is that if an enemy player is wrapping around you, it will accelerate towards them and start firing and may move the mouse slightly past them or to the edge of their body in the direction they're wrapping around you and then settle as a real player would (instead of locking onto the body that looks obvious)...but all of this with no user mouse movement required. Completely on autopilot for the situation one wants to activate the mode while looking real at the same time. This would be a gamechanger.

You can achieve looking legit with Smoothing anyway. Which will allow you to overcorrect or miss the target slightly. Also, you can set it so that all of this is done with 1 key input without any user interaction including shooting.

What you are saying "Just tell the AI to do what 1% Player does" means we need Data and tons of it. Not some shitty ass numbers either, but proper data on hardware to make it 1-1. None of this is worth any effort unless we have something to test with which we don't.

Until this AI thing blows up properly. We live in a time where 90% of anti-cheats can't even detect DMA at the moment, I don't see this happening not from us or anyone else properly. I say keep your excitement on the lower end.

AI vs AI will happen at some point. Not anytime soon.

[Wallhacks]

You can achieve looking legit with Smoothing anyway. Which will allow you to overcorrect or miss the target slightly. Also, you can set it so that all of this is done with 1 key input without any user interaction including shooting.

What you are saying "Just tell the AI to do what 1% Player does" means we need Data and tons of it. Not some shitty ass numbers either, but proper data on hardware to make it 1-1. None of this is worth any effort unless we have something to test with which we don't.

Until this AI thing blows up properly. We live in a time where 90% of anti-cheats can't even detect DMA at the moment, I don't see this happening not from us or anyone else properly. I say keep your excitement on the lower end.

AI vs AI will happen at some point. Not anytime soon.

Yeah, that makes sense. To train an AI that is good on its own you need a lot of data (and good data as you've stated). You'd need really good players to be training it ideally, or a blend of each individual user training it combined with the historical method of the aimbot code blended into one. May not be the easiest thing to do.