Steam Games Must Fully Disclose Kernel-Level Anti-Cheat On Store Pages

An anonymous reader quotes a report from Gaming On Linux: Valve announced a change for Steam today that will make things a lot clearer for everyone, as developers will now need to clearly list the kernel-level anti-cheat used on Steam store pages. In the Steamworks Developer post Valve said: "We've heard from more and more developers recently that they're looking for the right way to share anti-cheat information about their game with players. At the same time, players have been requesting more transparency around the anti-cheat services used in games, as well as the existence of any additional software that will be installed within the game."

Developers with games already on Steam will also need to do this, as it's not just for new games coming up for release, and it is also part of the release process now too. So Valve will be doing checks on games to ensure the notices are there and correct. However, it's only being forced for kernel-level anti-cheat. If it's only client-side or server-side, it's optional, but Valve say "we generally think that any game that makes use of anti-cheat technology would benefit from letting players know".

Who's Kernel?

[xaosflux]

Saying it doesn't apply to client-side installs makes no sense, that is exactly where the ring-0 concern is, on the client machine. Sustomers surely don't care if the devs want to install things "server side" in their own ring-0.

Re:Who's Kernel?

[jonwil]

What Valve is saying is that kernel-level anti-cheat must be disclosed but user-space client-side anti-cheat that lives only in the game process is not mandatory to disclose.

Re:

[jargonburn]

I believe "client-side", in this context, is referring to the game application itself as a client of the online server hosting the session. In which case, there would be no concerns beyond normal application security.

Kernel anticheat has been a boon

[fafalone]

Wherever there's anti-cheat, there's cheaters trying to bypass it. Cheat makers have moved into kernel mode too, and even though I don't make cheats (or even play any games besides a couple 20+ year old ones) these communities have been a wonderful source of information for learning how to make nontraditional drivers. I love making things a language isn't supposed to be able to make, and that's required learning a lot of details about Windows kernel drivers that simply aren't covered by books and classes on

Linux only it seems

[fjo3]

Windows 10 user. Just opened Steam, went to the store, and looked up a game I am 100% sure has anti-cheat software (Endless Dungeon). Didn't see any warnings. Oh well - yet another incentive to make the switch to Linux.

Re:

[drinkypoo]

There is EAC for Linux. Or you could buy the game and then download a hacked version that removes the anticheat. Or you could just not want to have kernel-level anticheat no matter what OS you run.

Re:

[SlashDotCanSuckMy777]

"Which games need to do this?
Going forward, when you submit a new game to Steam, if your game installs a client side, kernel mode anti-cheat, you will need to fill out this new field. We will be going through old games and contacting partners with games that fall into this category."

New games will have to do it, but games already on Steam will need updating.

Not sure the 3 hours this decision has been out will be enough time for all old games to update their game pages.

Re:

[thegarbz]

Why would it be Linux only? That makes no sense. Actually nothing you wrote makes any sense. Firstly the announcement only came out today which means few (if any) store pages would have been updated already. Secondly Endless Dungeon does not use any kernel level anti-cheat (e.g. Easy Anti-Cheat). Thirdly most games which do use kernel level anticheat have compatibility problems with Linux.

Kernel-Level Anti-Cheat

[Valgrus Thunderaxe]

Why isn't this considered like a virus? Why don't AV vendors alert the owners of these machines and try to remove this code?

Re:

[thegarbz]

Why isn't this considered like a virus?

Probably because it's not malware, not replicating, and not malicious. Why would AV vendors try and remove it? What next, AV vendors remove your display driver as well because it runs close to the kernel?

Re:

[evanh]

It is fair to class any form of DRM as malware. It's primary function is interfere with normal operations. As such it also tends to play havoc even when not triggered for that interference.

Re:

[evanh]

DRM has similarities to security back doors.

Re:

[sinij]

Unfortunately, because a typical user will pick lack of security over inability to play a game every time. Do not interpret this as me saying that AV vendors offer any measure of security.

Just say no to anti-cheat

[WaffleMonster]

Won't even consider running a game that installs kernel drivers for any reason.

Would be nice to let users filter out DRM games

[Improv]

I don't buy games with Denuvo. Would be nice if I could set something so I don't see such games in the listings. At all.