Can you stop using ChatGPT to write your responses? It’s really not helping you.
1 Like
I am not using it to WRITE my responses…
Sure you aren’t.
You just normally write using bulleted lists to summarize points, leading emojis to emphasize paragraphs, frequent bold text, and em dashes.
Your buddy seems to disagree with you after I sent him the link and asked if anyone is using AI in this thread 
But while Tephra picked it up almost immediately based on style, to me what made it evident is the typical “moving the goalpost simply because new information was provided”
See? now you all of a sudden “know” that the blueprint code example I provided is first of all a blueprint code and secondly actually solves the issue. I provided new context for whichever AI you’re using and they suddenly started factoring in information they didn’t consider.
However as for the point you raised:
Says who? 
players have a movement component that contains their actual velocity vector.. which can be compared to allowed max speeds and ping is also readily available and you don’t need an AI to simply account for a pattern.
Edit: oh and one more note which does make it funnier..
Do you know what is another massive red flag that you’re using AI?.. This:
I have yet too see a single Conan player who posts something on the forum, gets called out that they’re wrong and actually concedes.. Usually they either argue endlessly even after they realize they’re wrong and make you feel like you’re talking to a wall (see exhibit A) or they will “silently disappear”.
OMG you are wondering that someone can agree that someone right?
Yea lol.. it’s a bit sad, but it really doesn’t happen, that’s why I included it as an extra joke You’d be surprised if you looked around the forums a bit. It’s savage like hyboria
PS And sure only AI can started consider some new information… This is inaccessible to humans.
Congratulation you are meet here some another kind of person who can do it…
If necessary, I could even insert a graph into the text. Can you imagine?
Can you imagine, someone else besides you knows what a blueprint is?
Apparently you didn’t notice he said, “included it as an extra joke”.
Even ChatGPT recognizes jokes, you must be using Grok for your posts.
1 Like
Sorry but no, it’s not just simple things. In fact the more complex the thing is, the more important it is for (lots of) human oversight to review anything AI does. The more complex something is the more often (and more significantly) AI will “hallucinate”.
- AI is only useful if you don’t care about errors, because you’re going to need a human to fix them.
- AI is more useful for jobs that produce static results (like writing a term paper).
- AI is less useful for jobs that are dynamic or ongoing. Anything that runs 24x7 can be done better by building the code into the program and/or even by running scripts outside of the game that monitor the game logs.
A) Use cases: good and bad
If you want AI to generate 10 different formats for your resume so you, the human, can compare them and pick which one you want (after manually editing it to fix the mistakes).
For this, AI is good.
If you want AI to generate 100 different pictures from the prompt: “Monkey juggling goats in the style of Salvador Dali”, and then have a human pick the best one for their dank meme.
For this, AI is good.
If you want AI to write a paper on the Battle of Gettysburg, and then a HUMAN edit it to fix mistakes, so you can cheat on your history assignment?
For this: AI is good.
But… if you want AI to monitor a bunch of log files, correlate the data between those logs to find criteria that might represent cheating, and then have HUMAN review all of the reports, so that a HUMAN can log on to the server and see if they can catch people cheating?
For this: AI is bad.
B) Cost
And we haven’t even talked about the cost of AI.
The compute power and the electricity required to run AI are insane compared to current, state-of-the-art tools for logging, monitoring and alerting. You can do these tasks more efficiently and much more cost effectively by building them in to the code or using existing, well known tools.
For example, having scripts that run on the game server (or even on a single server that monitors many game servers) is MUCH cheaper and uses MUCH less electricity than trying to have an AI do it. You can run bunches of shell scripts in crontab much more efficiently and cost effectively than an AI agent.
C) tldr;
Logging, monitoring and alerting is a set of jobs that can be done better (and much less expensively) using scripts and other existing automation tools. AI is basically the worst choice for this job.
D) Where it can help
The one place that AI could be useful would be helping people write those scripts and tools they’re going to use. AI is a good way to speed up the processes of developing code and writing scripts, but the code it creates always needs lots of review by humans and is only as good as the person using it. You have to be good at figuring out how to prompt it and how to review, troubleshoot, fix and validate the code that it creates. Writing code is not as easy as writing a term paper, typos in code can be fatal.
So again, it’s good for static tasks, like writing a script, just like it’s good at writing term papers. “Write a script that will do A, B, D and D”. Then, after you fix the script it creates, you can put that script on a server and run it 24x7.
But what you don’t want is to have the AI do the monitoring.
We all sympathize with your desire to see less cheating, but AI is simply not the right tool for the job. Don’t believe the marketing hype.
1 Like
Conan Exiles servers
Isso não causa LAG o Battleye funciona normalmente, mas para quem estiver com problemas configure o firewall
Well, here I am complicating everything, while Xevyr suggested a much simpler solution that doesn’t require a huge amount of resources and time in his blueprint example that he shared above. That’s all great, I agree with it and fully welcome it. BUT in that case, even more questions arise for the developers. If there is a simple solution that doesn’t require massive resource costs, where one person only needs “15 minutes,” but most importantly, it only needs to be done ONCE and it could solve the problem FOREVER, then I’ll repeat myself: this raises even more questions for Funcom.
Maybe you didn’t read it very carefully, but the “extra joke” was about something completely different.
Well, that comes down to what I said above pretty much.
They have a pretty long list of actual bugs that occur with the native features of the game.. @Kikigirl has been keeping quite good track of them too on here in case you want to check those out 
here’s one example thread
As I previously mentioned they are also running Conan on limited manpower atm. They keep referring to them as “team”, but honestly I am fairly sure it’s a single developer on it and might not even be full time - though this is my speculation.
If we put the above two things together, you’ll realize that even if there was a “simple fix” (more on this below), the bugs that are actual parts of their game most likely take priority over fixing deliberate interference via external software.
Now whether everyone agrees with that is ofc another matter, but at the end of the day, it’s kinda like.. If a restaurant doesn’t have food, the fact that the music is bad becomes a secondary issue.
As far as simple fixes go, sure, a lot of things have very simple fixes that can be done in 5 minutes or so each.. The issue becomes making sure that it’s indeed the correct 5 minute fix and for that the developer needs to look at the grand scope of things.
For example I already mentioned things above that would require my simple solution to be expanded. like the fact that it would prevent quite a few mods from working all of a sudden unless they add a setting or figure out a way to interact with them better as a lot of them rely on “remote looting” or pulling items from chests to craft etc.
This might make the proper fix quite a bit longer than 5 minutes once they finish considering all the other systems it could potentially effect, sometimes tiny changes have quite serious effects in systems that don’t even seem related, which needs consideration.
Afterwards they would also have to test these changes and catch any issues with it in time to come up with a “fix to the fix”..
Overall, what might take me 5 minutes to casually throw together (even I made a mistake initially and quickly replaced my original picture after posting) takes probably just as much if not less to the actual devs, the problem is that they need to follow proper procedure and all of it needs to go through the whole process, which simply makes even simple things take much longer.
Hopefully that explains it a bit better.
2 Likes
What you wrote is perfectly clear to me. But in my opinion, I would assign a much higher Severity and Priority to cheats than even some bugs.
An uncomfortable seat in a cinema might not drive visitors away as much as mentally unstable people who come in and start causing chaos, ruining everyone else’s experience. Compared to that, the uncomfortable seat becomes a minor issue.
It’s the same in the game: when you log in and find your base wiped, all your thralls killed, and on top of that all your knowledge reset (and as we know, some of it can only be obtained through events), then a bugged inventory feels far less discouraging in comparison.
I believe many players would agree with me on this.
2 Likes
Wow, Grok has even less of a sense of humor than I imagined.
1 Like
Because of a combination of the following reasons:
- Conan Exiles is very low priority for Funcent right now, so they won’t invest a single penny more than absolutely necessary, which includes paying people to look at reports.
- Investigating some of these claims is easy in terms of complexity, but time-consuming without proper tooling.
- Investigating the rest of the claims is very hard and would most likely require adding more logging, instrumentation, and monitoring.
The only aspect of these problems where you could shoehorn the AI you’re trying to peddle is the tooling and @Xevyr already pointed out that you don’t actually need AI for that.
The fact that you think AI is a magical solution for that goal tells me you don’t know enough about it to be worth listening to.
And this tells me you don’t know enough about game development in general, and Conan Exiles specifically, to be worth listening to.
AI is not magic. You can’t just wave your hands at it and say “detect weird patterns” and expect it to make the investigations faster.
Actually, it is. I suspect we’re dealing with someone who has a hard time expressing themselves in English, based on the incomprehensible quality of the few posts where they didn’t use whatever AI tool they’re relying on normally.
2 Likes
Actually it’s the other way around, you’re significantly over-simplifying everything.
It’s a simple concept, and it’s easy to say, “Let’s use AI to fix cheating”, but just because something is easy to say doesn’t mean it’s easy to do. The problem is complicated and difficult to fix. If you don’t agree with the argument that the problem is difficult to fix then spend some time looking at the forums of any competitive game.
Every game that features competition between players, and I mean EVERY game, has a cheating problem, and every game has lots and lots of discussions about how to stop cheaters. It doesn’t matter whether they’re FPS games (e.g. PUBG, anyone?) or RPG’s (ARK, which is a lot like Conan, and is a wretched hive of scum and villany), MMO’s, or anything in between, cheating is a rampant problem for every game. Even games that are super casual, e.g. “cozy” games, are constantly battling against cheaters who are either cheating just for the fun of it or becasue they want more loot for the time they spend playing the game.
It sounds simple to suggest that AI can help fix cheating, but if that was true then the game industry would be bragging to anyone and everyone who would listen that this is what they’re doing. They would be bragging to their shareholders that they have found a cost-effective way to use AI to stop cheating and boost profits. But they’re not doing it, because even the best anti-cheat solutions are neither easy nor cheap.
So you’re not “complicating everything”, you’re vastly over-simplifying it. You’re misguidedly arguing that there’s a simple and cheap solution when that’s not true.
Again, you’re over-simplifying. Even while he was explainting it, he also pointed out that there are additional difficulties when actually implementing the solution in the real world. Even Xevyr’s “simple” solution isn’t simple actually make happen (and he didn’t pretend that it is). What he really said is that it’s less complicated and more effective than using AI.
I don’t intend this as an insult when I say, no, it raises questions for you. You clearly don’t understand the difficulties and complexities of either the cheating problem nor of the difficulties and complexities of creating solutions. That’s pretty understandable, it’s a very common mistake for people to look at an industry or profession and underestimate how difficult and complicated it is. Every job seems easy when you’re looking in from the outside. (Remember this conversation when other people try to tell you how easy your job is, and you want to roll your eyes at them).
ALso, the AI industry is doing everything they can to convince people that their product is so smart it can do anything. Don’t believe them, don’t believe their hype, don’t believe their marketing, don’t believe their executives, don’t believe their humble-brags, just… don’t believe them about anything.
3 Likes