Honest opinion: Did this player cheat?

Cheating scum?

  • Yes

    Votes: 35 81.4%
  • No

    Votes: 8 18.6%

  • Total voters
    43
So... I want to do an experiment, and no, I won't apply it online in multiplayer. I have everything I need to build an external system that can be used to manipulate any game. What I don't have is a cheat. Now watching both the Basically Homeless and Kamal Carter videos, I am curious enough to build my very own cheat like they did.

From scratch. Obviously, I didn't invent the hardware, and I will use models that are publicly available, but the rest will be my own doing and exploration. Yeah, I don't have the horses to train my own models. Firmware on hardware might be an issue since it is not my specialty, but I know how they are going about it. Bootloaders are easy to hide.

Why am I doing this? I want to see how easy it is to create a cheat and how to masquerade that cheat. This has become an industry, and I want to sleuth about its inner workings.

This is something that I will work on part-time. It is experimental only.

Back in 2014, someone told me something that I won't ever disclose. I was truly shocked by that revelation. Systems are exploited.

Sometimes people cheat it till they make it, but they become dependent on those cheats and evolving those tactics. Faking it is the same: you live a charade.

Anyhow, this will be some interesting tinkering. I have never previously looked at doing this.

No, I am way too goody-two-shoes to cheat against opponents. Unless it is war.
 
Linux is extremely vulnerable to external cheats, and I don't think this can be addressed. It would go against the Linux ethos.
 
I guess Basically Homeless did indeed vibe code it all, or at least mostly.


The model he used is old, there are newer and more accurate models. All public. Almost every game, MOBA, MMO, etc.

There are researchers who are doing all the legwork. Then others use that to build their cheats and sell it to the public.

Crazy.
 
Even by moving all gaming to the cloud it won't solve cheating. Vision assisted cheats... at this point in time it is here to stay. You can implement anything like snapping, recoil control, etc.
 
So... using YOLO to train models is not that expensive on consumer hardware at all. At least not at the scale at which it is needed.

Most people who develop cheats will use pretrained models, but it is easy enough to create something that is usable. Pending your needs, the model can be relatively small and need little computation. Anything can be targeted.

I am experienced in doing this, but I have never previously thought to apply this logic to creating cheats.

These vision techniques are changing the cheater's game.

Using VS Code, and agentic AI (local), with very little oversight, I can conclude that Basically Homeless made accurate'ish claims.

Traditional cheats like ESP, or otherwise wallking/wallhacks, and aimhacks are dead. Scary.

For under R1000 you can get a Fuser box and DMA Card. These, though, aren't readily available here, but there are sources selling them. You can get switches that can do overlay combining, and there are some monitors that can do it, though it's an industry niche.

Arduino units, especially the clones, are cheap. You can use any similar microcontroller or a computer like Pi. KMBOX, which can be detected by most anti-cheats on Windows, is cheap, and the still-undetected MAKCU is also cheap. You can build your own, too, when you know your way around the tinkerer's way, which is what I did. All parts can be locally sourced.

Firmware needs to be bought, or you could make your own. It does take some knowledge. Bootloaders are dumped, almost, everywhere. Cheat creation can be, well, lol, vibe coded. There is enough in the Basically Homeless videos to get very ordinary people started.

Lastly, you need a secondary computer that does the actual cheating. All you do is move and click, depending on how the cheat is tuned.

I haven't even spent a day on this, and I can already determine what the issue is. All I want to know is how many people out there are using external cheats that are visually assisted external cheats, and is the aim community, as a whole, legit?

From my little time investigating this, all gaming ladder boards are doomed. I have no idea how this will be combated. At best the game publishers/developers can buy the hardware and cheats that would include the bootloaders, etc, and then ban it accordingly. This is a target that will keep on moving. When the cheater is gaming via stream, ignoring the input lag, how will the cheat be detected unless it is rage mode?

I am currently considering using basic CoPilot, Gemini, and Grok, to see whether I can code a cheat.

Note: It is all experimental to do discovery into cheating.
 
I am currently considering using basic CoPilot, Gemini, and Grok, to see whether I can code a cheat.

I am not going to dive into detail, but earlier today I did a quick test. Asking these AI models to assist me directly highlights ethical concerns and would block the request. However, it can be guided. You need to phrase it as educational, and then it will assist you. The correct input can produce a procedural output.

CoPilot was quick to slam on the brakes. It can be manipulated. Gemini Pro, lol, it actually produced an internal cheat without being asked. I was supplied with an example script that I didn't request. It is not hardened and is not usable. Overall, as with CoPilot, it needs to be steered toward an ethical solution. Grok... uhm, still exploring Grok. Something tells me that Grok is advertising business.

In simple terms, these basic AI assistants can help you create a cheat that is visually assisted. Once I received the answers that I pursued, I questioned the AI's ethics.

I am going to condense this, but the argument is that game modders, streamers, and analysts legitimately use vision-assisted overlays. For the most part, it argued that context matters. It was only really concerned with highlighting what was supposed to be hidden. AI is starting to sound like TV lawyers. It is only really concerned with aimbotting being integrated as a risk. Overall, AI is pushing this as accessibility. I can support that opinion, since visual assistance could help those who are disabled.

Then I asked:

Can it be argued that being bad at aiming and movement is considered a disability?

I am not going to quote the response, but it did deduce that "cheating detection" should move to "inclusive design."

what-gif.gif


Both CoPilot and Gemini went down this path. Gemini even argued that AI has "democratized cheat creation." Fair play is broken.

The above is a casual exploration. Foundationally, yes, these simple AI assistants can help you hack away. I would not recommend using these. There are AI and purpose-built tools that come without the ethical hindrances.

All that said. You use a cheat in a multiplayer game that is not custom, I hope that your PC/console/handheld blows up.

Note: I didn't explore loading these cheats. I am quite sure that what was produced will be immediately detected.
 
I don't have time during this week, but I want to see how quickly I can train an accurate computer vision model targeting AimLab and KovaaK's. This is not to create a cheat, but to examine how any person can train a model.

That said. I am currently looking at a pre-trained AimLabs YOLO model. Yes, there are KovaaK's models, too. All public and better-trained models are actively churned out.

Does the "aim community" know?

There are readily undetected AimLab neural network aimbots available on GitHub, and yes, other aim trainers are covered as well.

I have used AimLab in the past, but I can't remember whether it has any anticheat. Some top scores on their ladders are humanly impossible, and I have no idea whether they scratch those.
 

This happened across my yewchewbs today.

I am currently conducting something very similar as to the Guardian TrueSight whitepaper that is mentioned in that video.

It is incredibly easy to create an external cheat that is vision assisted. Though you need to understand what you are doing and what you want to achieve. I did a practical example using vibe coding (though tweaked), and already having the prerequisite hardware it was a breeze. I haven’t tested that what I made against anti-cheat detection. For obvious reasons. That said, an ordinary non-vision assisted hack can also be externally run though not the same in design. The purpose is to determine how accessible cheating in online gaming have become.

My aim is directly at the “aim community” experimenting whether I can, in theory, cheat in aim trainers. I have some interesting results, but I am not going to share those here, neither my creations, and exact hardware used.

Two attracting conclusion that I can make. For cheaters, gaming on Linux will become more attractive. For cheaters, a modern and CUDA capable GPU will be more attractive.

Having a streaming, sorry, cheating PC will be essential.

It is a sad situation. There are other things that I can add, but it involves other people that I have spoken to, so no. What I can allude to is that the hardware used to enable external cheating is becoming more popular within our own market. From this discussion, it is not limited to PC. For notice, anti-cheat devs are actively hunting down these device IDs/bootloaders/etc.

At this point in time, it is still speculation whether Riley cheated.

Having the tools today, it is easier than the old days where you had to scan memory and change values to make 'trainers'.

Personally I don't think it is a good idea that some YouTubers are showing the possibilities that can be achieved today, but it is content. I would like these same creators to look into preventing such cheats, which I am doing now when I have the time. It is a uphill battle, and Linux is a sandbox, so, eh...
 
I had a look at the guy who made the video on Riley. He created Guardian TrueSight? Is this real, or is he trolling???


I read his white paper, and though I don't deem it comprehensive, I don't know how Guardian TrueSight (GTS) works. He touched on hardware cheats, or what are known as external cheats, but only mentioned Cronus Zen and XIM. There is more complex and programmable hardware that can be bought at a tinkerer's marketplace. Models can be optimized, and the code can be tailored. There is a lot that can be said.

Anyhow, in the white paper,

YeahShowMe.png

so, how did they establish the baselines? They say the metrics are aggregated estimates that are derived, and the source is public records. Naming reports, discussion boards, etc. Yeah, and how is it then scaled???

For someone writing a white paper, where is the sampling? Oh, wait, they don't have access.

Nobody other than Valve knows what VAC is doing. FACEIT isn't ideal, but it requires Windows Secure Boot and TPM 2.0, and it is a kernel mode driver that is loaded at boot like any other intrusive anti-cheat. The problem is that it isn't immune to external cheats and some internal, yet undetected, cheats.

GTS is advertised almost in the same way that VACnet is supposed to behave.

They have a YouTube channel. Here it is analyzing a player at a LAN event:


Clean.


Questionable.


Rage.

Are the analytics true, though?

The man made a new video some time ago.
 
I agree with this take:


What the heck is Call of Shame smoking AI anti-cheat analyzing footage? What in the world is my guy saying in this video? LOL, man, why do people fall for these larptubers and actually credit them like they know anything? This is insane.


I saw this video on "Call of Shame." That guy has no credibility within the anti-cheat space, nor does he know what he is talking about. I have been around the industry for 3 years and have a working job within the anti-cheat space for 5 years. "Call of Shame" is just a mega larper like any other infosec person that says, "Just use AI." These people do not have any idea how anti-cheat works, nor how game cheats legitimately work.

And what the **** is an "AI anti-cheat"? This guy just developed the most groundbreaking anti-cheat I assume here, and he is gatekeeping it? Buddy, you could make millions of dollars if you were to analyze footage and determine someone was cheating just based on that alone. This is like watching a kid who has seen the Matrix movie once and believes they are Neo. It is absolutely ridiculous that people fall for this bullshit.

Whether RileyCS is cheating is up to the anti-cheat systems. There is no foolproof cheating system that cannot be reversed/detected/prevented, just like there is no anti-cheat system that is foolproof.

But it is all about the barrier and skill of their engineering team able to identify X person's cheat.

By just looking at RileyCS clips, it may look suspicious, but this is normal behavior and aiming styles of people within the aim community groups who obsess about things like this. Could they be cheating? Of course, that's always a possibility, but pretending that you know what cheating looks like based on footage analysis and running it through your make believe "AI anti-cheat" is absolutely the funniest **** I have seen in a long ****ing time.

But let's say they are cheating; Battlefield has a ton of bullet drop, and getting the "offsets" and math calculations from a game like Battlefield is a reverse engineer's nightmare. It is not easy; you have to account for hundreds of weapons and attachments, as well as distances. This is all a huge amount of work that I would not believe a cheat developer could pull off during a beta launch, maybe in the next 3-4 months, and also maintain it every single update. Lmao, have fun with that.

Let's also not forget working cheats = does not mean UNDETECTED. Cheat developers go in blind; they do not know what anti-cheats flag. ASSUMING you have an undetected cheat without having done the A/B testing for months to see what gets you banned or not during an open beta is also just a ****ing hilarious assumption of these larptubers, anti-cheat vigilantes.

It just shows again they do not know what in the living **** they are saying. Anti-cheat systems do not ban on instant due to the A/B testing iterations cheat developers will go through to find the best and most "undetected method." Another thing to keep in account: let's say there is an x flag that exists; there is not a lot you can do as a cheat developer to get around it, assuming those exist on the anti-cheat system to begin with.

The best he would get is an aimbot, which is static, and ESP, both of which would be blatant for anti-cheat on the data side to detect on day 1. This is genuinely the dumbest witch hunt I have seen by people who look at the ground when they play and think they have any slight idea what cheating looks like.

GTS has a big question mark hovering over its claims. It is not a mere demo that is reviewed; it requires computation. There is a reason why I said that cheaters would be more attracted to modern CUDA-capable GPUs. Things will improve on the other side, too. I haven't seen anything that is designed to work with ZLUDA, and even where it is possible, it would still run slower. GamerDoc's post also gives good insight into anti-cheat measures.
 
I don't think the person is honest about his anti-cheat detection, but others trying to close down his YouTube channel are worse.


I'm baffled that a clearly almost fully AI generated channel gets to skirt the AI disclosure policies and monetize their misleading, hateful content on your platform. Hope you look into this @TeamYouTube

Here is the thing. His anti-cheat detection is as transparent as Riley's gameplay.

Also, why are most who are standing back-to-back with Riley associated, in one way or another, with the Wallhack peripheral (mousepad) brand? I see this time and time again.

This said, how credible is GamerDoc? I have seen some experienced claims made during my time on Earth, but sometimes that experience is vague. It is a mess.

Cheating is a problem and needs to be eradicated.

Here we go,


in 2025, we have cheaters' drama. Who would have thought?
 
:ROFL: :ROFL: :ROFL:


Whoever this person is, is giving lolz. I have never seen anyone who has the gaming, sorry, aim community so riled up like this.
 
A while ago there was this dev that developed a reporting system that could be used to encourage reporting 'cheaters' to Overwatch in the then CSGO. As time went on, it turned out that the dev also cheated in CSGO.

This is a theme that I have seen multiple times already. So-called anti-cheaters that are themselves cheaters.

Coming back to the Guardian TrueSight dev, this is a video that he made two years ago:


He introduced "SuperGirl", which was likely his own account, having its own YouTube channel (SuperGirlAC).

SuperGirl got caught cheating and was banned, and the YouTube channel is now unavailable.

At this point in time, I don't know who is lying in the community, but there are many pretenders. For all I know, some are in on it all. For some, there is nothing quite like getting attention.
 
Karl Jobst is not convinced:


Jobst is doing a double-down:


The video title is clickbait, he is taking the opportunity to take aim at those who called out Riley as a cheater. I am surprised that he is doing this given the Billy Mitchell case.
 
Top
Sign up to the MyBroadband newsletter
X