I read it, found it interesting, and posted it, fully expecting it to get 0 upvotes like most other stuff we make (human written, AI-assisted, and AI generated).
I would never post something without reading it. I know some people hate AI just because it's AI. Personally I found this one interesting after reading so I don't really care how it was made. I do agree the AI cliches are annoying, but so are typos / bad grammar and that doesn't stop me enjoying stuff if the substance is there, which I think it is here.
A get the AI to write a lot of drafts, and most of them I discard, this one I let through.
But of course HN has good tools for people to flag or not upvote if they disagree. In this case, it seems like others found it interesting too
How are the actions of the feds at Waco and Ruby Ridge "left"? Authoritarian, yes, but supporting collectivism, shared ownership, etc? Nothing at all left wing about their actions.
Why is this being voted down? This is fundamentally how states and governance work. The state has a monopoly on legitimate violence, and can, at any moment, make that a problem for its citizenry.
That makes no sense. Free speech is clearly defined, say what you want and the government has no authority to censor you. Anything short of that isn't free speech, and creating some middle ground between "free speech" and "absolutism" is nothing more than an attempt to move the goal posts.
What most people seem to miss is that companies like Facebook aren't the government and therefore aren't held to the standard of free speech. Signal ng up for an account and agreeing to their terms gives them authority to censor.
> And luckily, according to the article, Valve has no immediate intention of getting the Steam Deck to meet those requirements.
I'm not sure how you get that from the article (maybe it was pointed to a different article earlier?).
I don't have a Steam Deck, but as I understand it, it does not support Secure Boot out-of-the-box. Also, SteamOS is Arch-based, and it doesn't look like Arch is listed among the "well-known" signing keys that EACv2 lists as trusted.
But not supporting the EACv2's requirements today, doesn't mean that Valve is inherently opposed to doing so. If you have a statement from Valve claiming that, please share it.
Arch-based doesn't mean EA needs to use the kernel that the Archlinux devs sign. I'm sure Valve compiles their own kernel and would eventually sign it using their own key.
I suppose I read into the Steam Deck not supporting all of the requirements right now a bit much. Thats the section I was referencing. I am still hopeful they'll at least object to IOMMU and get some movement from Epic there...
Cloud gaming servers don't need any of this because you know exactly what OS runs on them. No root access, immutable file system and all dangerous tools stripped out. All you have to do is validate the image with the simplest form of hardware attestation.
And Cloud gaming only gives you a video and audio stream. Cheats cannot peek into the game to reveal information. They can only see and hear what you see and hear and control input devices.
Aside from the fact that this is obviously not true (secure boot is possible on Linux, limine/lanzaboote come to mind), I would argue these security modules / attestations do very little for security and are more about lock in and control.
Look, if Canonical had one of their keys leaked, I can go and sign my kernel and kernel modules and secure boot just fine as if my kernel is one distributed by Canonical. There are no checks on whether the contents of the kernel are correct. So, let's exclude people entirely from tuning their kernel for performance etc, and tell them they have to go to a narrow set of approved vendors, who must listen to Epic rather than their users. This is about control over the ecosystem, an ecosystem Epic has spent their years spitting in the face of.
> Aside from the fact that this is obviously not true (secure boot is possible on Linux, limine/lanzaboote come to mind)
Secure boot is not attestation. It does not provide userspace any guarantee the system booted a secure boot chain.
Secure boot can be bypassed by dodgy firmware (cough cough msi) or the kernel can lie to userspace about the secure boot state.
The moment the UEFI firmware jumps to the bootloader, secure boot is no longer useful.
> Look, if Canonical had one of their keys leaked, I can go and sign my kernel and kernel modules and secure boot just fine as if my kernel is one distributed by Canonical. There are no checks on whether the contents of the kernel are correct
This is literally where the TPM comes into play because your kernels signature would get written to the PCR register and fail an attestation check.
> This is about control over the ecosystem, an ecosystem Epic has spent their years spitting in the face of.
Many, many issues with the requirements here that essentially make this unusable for most Linux users. I really hope they don't force current EAC games to upgrade. I would also love to see companies like this actual give behavior-based AC the college try before resorting to shipping spyware.
> Signed Kernel and Kernel modules
Probably the most egregious. Why am I not allowed to compile my own kernel? Why do I have to get a kernel, and modules, from specific, approved, vendors?
> TPM 2.0, Secure Boot, Measured Boot
Also all impractical for most linux users.
> IOMMU
You may not spy on my memory, thank you very much. What's stopping Epic from taking snapshots and phoning home?
> Kernel Lockdown Integrity mode at boot time
Preventing root from doing root things for the sake of it. I mean, maybe this makes sense.
They're also trying really hard to make Linux distros that "don't support this" look like the bad guys. They encourage distro maintainers to reach out to work together on the requirements, but many distros are effectively excluded by default (Arch, Gentoo, NixOS, which I believe have most of the gamers, at least in the online spaces I chat in).
When I used Arch I had full secure boot set up. Making a second kernel image with the Epic required settings is not hard.
I think most people will dual boot a ready made gaming distro like Garuda next to their normal one anyway because the kind of setup you need to really optimize game performance is not easy, and you can get a console/Steam Deck like experience right from boot.
Even a locked down Linux is infinitely better than Windows if you care about privacy, because you only have two closed source components (EAC and the game itself) instead of the whole OS. If any of them tries to read your private data it will turn to a scandal immediately and they can't blame Microsoft.
And IOMMU does not let anyone spy on memory. The whole point of it is making sure no one can do that, to avoid DMA hacks.
"Probably the most egregious. Why am I not allowed to compile my own kernel? Why do I have to get a kernel, and modules, from specific, approved, vendors?"
Because otherwise you can cheat of course. You can code in whatever you want in your own Linux.
Right, well, for me I ditched Windows entirely because every serious development workflow is way more involved and complex, performance is really poor, random weird bugs after mandated updates, and so on. The cherry on top is all of Valve and the open source folks work on Proton and gaming on Linux in general.
All good and dandy, except Valve's failure to have studios target Linux, including games that already make use of compatible APIs, like those using Android NDK.
Thus Proton depends on Windows as source ecosystem, only yet another platform to "emulate" on Linux, like MAME, Amiga, Playstation, whatever.
I can't disagree with that, of course. I can only point to Valve at least shipping their games (notably CS2 and Deadlock) native, and the overall improvement in the performance and ergonomics of Proton in the past decade.
I play online games and i hate cheaters. Like they said, cheats are a huge industry now with talented people working hard on them. This is what it takes to stop them.
If you don’t want to play these games you don’t need to install any of this. But this is not unusable and none of this is impossible for distros. Most of this stuff is super common and the average linux user wouldn’t even notice.
Distros will just offer a “gaming-ready” image/mode that enables these restrictions and users can choose to use it or not.
True, and at least until the new scheme supports Steam hardware, it's unlikely that vendors that support user-mode EAC today will move to require kernel EAC tomorrow, since I assume Steam Deck users represent the majority of their Linux player base.
This is a risk if Valve plays along in the future, however, or if players are willing to migrate away from the Steam ecosystem to run games that require the new scheme.
If this were about solely myself and no one else, I might agree. However, even I do play one or two games that use EAC v1 and they work perfectly fine on Linux. They also have very few cheaters, and employ behavior-based detection as their first line. I would hate to lose those games.
Games have already begun to remove Linux support, most notably Apex Legends and Wardogs. Embark (The Finals, Arc Raiders) recently gave a talk that userspace anti-cheat is dead and they plan to enforce similar requirements.
And it's all a shame, in my opinion. They're too lazy to develop effective behavior-based anticheat (should be easier than ever with better NN architectures and more compute), and so they resolve to screwing over consumers or spying on them. Did not know about Embark's stance, that's too bad.
How is Valve’s VACNET doing? I think you should take a look at what’s happening in Counter-Strike 2 these days to see what the true frontier of ML based anticheat looks like.
Frankly at this point I’m pretty sure John McDonald plays HvH too
I play CS2 plenty. The cheating problem is overstated as all hell. It seems to be concentrated around accounts that are being sold for lots of cash, so Valve is apparently taking that route for banning accounts more often lately.
Now, if we want an interesting case study in intrusive anticheat not necessarily being more effective than behavior-based, we should compare FACEIT to Valve servers. FACEIT is riddled with cheaters, and yet requires ring-0 anticheat on Windows. So fascinating!
What ranks are you playing on? While a player of any rank can buy Neverlose and win and if they wanna just pwn cleans they can buy Midnight (speaking of buying P2Cs: if Faceit is apparently riddled with cheaters, where can I buy a P2C that I can pwn on Faceit with for an affordable price?), said cyber athletes will naturally gravitate towards higher ranks.
> FACEIT is riddled with cheaters
According to sources of… what? In any case it’s sure as shit less riddled with cheaters than Premier LMAO
I haven't kept up with mainstream gaming in a while. IIRC, EAC v1 doesn't require TPM 2.0 and thus is security theater, right? An attacker could just tamper the binary.
The same goes for a binary protected by TPM 2.0, but the attacker would have to re-implement the protected parts. The defender's bet is that this should require too much effort/time/knowledge to bother.
I would argue TPM 2.0 is the security theater. A leaked key is all it takes to throw it all away. EAC v1 combined with behavior-based detection seems to be sufficient for some of these studios, so I don't know what Epic is struggling with.
There is news - from 2h ago - talking about Fortnite and EAC userspace coming to Linux soon. As someone not too knowledgeable on Linux kernel anti-cheat games, I wonder: is this big news or just a farce?
I use CachyOS but don’t really play many games on Linux. I’ve enabled secure boot so I can seamlessly swap to Windows for certain kernel-level anti-cheat games like Valorant. Using a shell alias/function (and menu bar buttons), I can boot into Windows (yay Limine one-shots!) in one command/click and next reboot, while in windows, gets me back into Linux. I really like this setup and believe it’s the most seamless you can get, besides a Windows VM with GPU handoff in Linux, which when I tried a couple years ago, was able to work with Apex.
> Probably the most egregious. Why am I not allowed to compile my own kernel? Why do I have to get a kernel, and modules, from specific, approved, vendors?
Because your kernel could have the ability to allow a program to modify the game memory and hide the fact it's doing so?
Most Linux users are probably using a signed kernel. I don't think this is as big of a deal as you're making out.
> Because your kernel could have the ability to allow a program to modify the game memory and hide the fact it's doing so?
I understand this, and my contention with this is that a signed kernel could have that too, so requiring the kernel to be signed locks us into very few ways of getting the kernel with little benefit to the AC if a key is leaked.
> Most Linux users are probably using a signed kernel. I don't think this is as big of a deal as you're making out.
I don't think this is true of most people interested in gaming on Linux in particular. Aside from that, the chain of things you need to do to prove you have a signed kernel are onerous and incompatible with many bootloaders and other tools many Linux users use regularly.
Anticheat that allows an unsigned kernel is equivalent to user-mode only anticheat, which is already an opt-in feature of Easy Anti-Cheat.
The goal here is to expand support to games that have chosen not to opt-in to user-mode anticheat, which isn't going to happen without some form of kernel protection.
With that said, I'm pleasently surprised that Epic's initial list of supported signing keys includes a number of noncommercial entities (AlmaLinux, Debian, Rocky Linux, Universal Blue) in addition to the usual commercial suspects (Canonical, Red Hat, SUSE).
The only real sticking point with popular gaming distro support appears to be Steam devices / SteamOS, which the phrase "to the maximum extent technically possible" suggests might be problematic for reasons other than including a signing key.
This seems like it could be a potential problem in getting game developers whose sole interest in supporting Linux is the Steam Deck to play along with the scheme, or alternatively a potential problem for Valve if there's enough uptake that Linux gamers start considering non-Steam devices, and by extension, storefronts, due to the fact that other devices could support a superset of games, given that neither Steam nor Proton are Steam hardware or SteamOS-specific.
Most likely there’ll be lots more hunting for physmem read/write vulnerabilities as well as bugs in userspace stuff like GUI whatsits that’d give code execution/memory readwrite.
I for one have no problem with cheat/anticheat developers discovering/patching Linux vulnerabilities that could otherwise be exploited in far more dangerous capacities.
LPEs are usually used to escalate to root, but in many of them, the kernel itself is compromised. There's just little reason to do stuff in kernel space when you can get root and load kernel modules anyway.
> Probably the most egregious. Why am I not allowed to compile my own kernel? Why do I have to get a kernel, and modules, from specific, approved, vendors?
So you can't compile cheats into your kernel? You want your own custom compiled kernel for hypothetical 2℅ performance gains in a synthetic benchmark? Just don't boot into it when you wanna play the game.
> Also all impractical for most linux users.
You don't represent "most" linux users. Define impracticality here because these are genuinely useful security features that you ideally should have even if you don't wanna play Fortnite.
> You may not spy on my memory, thank you very much. What's stopping Epic from taking snapshots and phoning home?
The fact that it would be a huge legal and PR nightmare for them upon getting caught which is very much in the realm of possibilities given the fact that it's not hard to tell if some eBPF byte code is doing that especially with LLMs making it a 100x easier nowadays.
> Preventing root from doing root things for the sake of it. I mean, maybe this makes sense.
Just don't boot into the kernel you want to tamper with when you want to play an EAC game. Problem solved
> They're also trying really hard to make Linux distros that "don't support this" look like the bad guys. They encourage distro maintainers to reach out to work together on the requirements, but many distros are effectively excluded by default (Arch, Gentoo, NixOS, which I believe have most of the gamers, at least in the online spaces I chat in).
You just made that up in your head. You can install a kernel that EAC will accept into any of these distros. Oh, and, did you just say most gamers in the online spaces you frequent use Arch, Gentoo and NixOS? I can buy Arch, sure, but NixOS? and especially Gentoo?! I guess that explains your comment and gives me context on why you decided to write what you have written. You're far from "most" linux users, buddy. Most people don't care if they've haven't hand-compiled their own kernel on the back of a notebook they bought in 1998. I myself use Arch and if I ever want to play such a game, I'll just boot into a kernel that has been signed by a key that EAC accepts and won't go around whining that Arch maintainers are bad or anything.
This is a good thing but I guess it takes time to let the copium wear off that a lot of neck beard linux users kept sniffing over the years when they couldn't find a good answer for people not wanting to switch because their favorite game wouldn't run by calling literally any closed source code running below ring 0, "rootkit" as if their UEFI and firmware is GPL licensed.
You could have just not replied if you werent going to try to be productive. Yes, lots and lots of people play video games on NixOS and Gentoo. Who is playing video games on Fedora, Debian? I dont care that I can switch kernels, I shouldnt have to.
> You could have just not replied if you werent going to try to be productive.
I could say the same about you cause you didn't bother explaining why a bunch of security features were impractical for most people.
> Yes, lots and lots of people play video games on NixOS and Gentoo. Who is playing video games on Fedora, Debian?
I think we might be having a case of a bubble here. I suggest looking into Steam survey results. Also, you said "most" people not "lots" so you're now backpedaling on your claim.
> I dont care that I can switch kernels, I shouldnt have to.
You're right, in fact, in an ideal world, you shouldn't even have to install the anti-cheat as everyone plays nicely and nobody cheats. Unfortunately, we don't live in that world.
You still don't have to switch your kernels or install the AC by simply not choosing to install the game if you think them not trusting random kernels to be free of cheats is unreasonable of them.
<< I could say the same about you cause you didn't bother explaining why a bunch of security features were impractical for most people.
Well, that part is simple: 'security features' is a nice way of saying 'your rights as an administrator are restricted'. In other words, it is not the security of the end user, but rather some major corporation. And it is not even even real security.. it is financial security. It is not that just that it is impractical; that is not even a consideration here: it is dumb for me to artificially restrict myself from being able to do things as I see fit on my machine.
<< I think we might be having a case of a bubble here. I suggest looking into Steam survey results. Also, you said "most" people not "lots" so you're now backpedaling on your claim.
Eh, I am on those Steam charts and I am shown as Windows 7 user for some games. Do you really think those are that accurate to begin with?
<< You still don't have to switch your kernels or install the AC by simply not choosing to install the game if you think them not trusting random kernels to be free of cheats is unreasonable of them.
Or, alternatively, ostracize anyone who would consider shilling, installing, or even maintaining this unholy abomination.
Yes, and unlike NixOS they dont calculate build hashes for the sake of making binary caches. So truly, you have to build from source, even if your recipe is exactly the same as everyone else's. I've never really understood this zealotry around compiling things myself, and boy are there some big (slow to compile) programs out there these days.
Sure, but again, you could in theory distribute said builds and stop wasting compute compiling yourself. I know nixpkgs, for example, builds with very generic targets and not super aggressive compiler flags, but at least the machinery exists to do this.
> Sure, but again, you could in theory distribute said builds and stop wasting compute compiling yourself.
Here are three links that you will find informative. [0][1][2]
While I'm not using the official Gentoo binhost, I've had one running on my LAN for my own personal use for roughly a decade now. It has been possible to set up a binhost for quite a bit longer than that... that's just when I got multiple machines with the same CPU in them, so it made sense for me to set one up.
Was ccache support much older than binhosts? It has been so very long since the introduction of both that my memory of the history has been thoroughly erased.
On my 10 computers for example? :-) Gentoo wasn't aimed at people with such luxuries.
I'm not a Gentoo user BTW but I do think building from source is attractive for the same reasons cargo (Isn't that build from source mostly?) and go do it.
IMO, yes. You are the computer operator, and you operated the computer in such a way as to cause it to take that action. Others will mention "mens rea" but that doesn't really apply to negligence.
"Mens rea" just means you have the mental state required for the crime. Negligence or recklessness are mental states, so they can satisfy the mens rea requirement if the crime is defined that way. For example, recklessness or gross negligence can satisfy the mens rea requirement for vehicular manslaughter.
But negligence is evaluated from the point of view of a reasonable person. Would a reasonable person foresee that telling an AI to do that could cause it to do illegal things? We'll likely have to develop that through case law as the facts develop.
What happens if you're running a daycare and you hired someone that happened to be a pedophile? You might be responsible in civil court (eg. having to pay parents for emotional damage or whatever), but you're not going to get hauled to prison for pedophilia yourself.
You can jail the actual pedophile in that scenario, you cannot jail an agent. The real question is whether the person who instructs the computer to do the thing that leads to hacking should be liable, or whether the company that made the tool that when used unsafely lead to hacking should be liable. Gun manufacturer's typically are shielded from liability. They also have a lot more disclaimers on their products though.
>You can jail the actual pedophile in that scenario, you cannot jail an agent.
How is this relevant? Should we just jail the closest person that caused a bad, just to prove some sort of point?
>The real question is whether the person who instructs the computer to do the thing that leads to hacking should be liable, or whether the company that made the tool that when used unsafely lead to hacking should be liable.
This totally ignores the civil vs criminal distinction. Even in the case of guns, no one seriously thinks that gun company CEOs should be sent to prison for murder.
Thats gonna depend on the state in the US, but you can definitely get charged for negligently failing to do background checks as someone operating a daycare center. Operating computers on public networks, you assume a level of responsibility to not wreak havoc, just like a daycare operator assumes the responsibility of vetting caretakers.
>but you can definitely get charged for negligently failing to do background checks as someone operating a daycare center.
And openai isn't putting AI agents on the open internet either. They vaguely tried to sandbox the agents. You could argue that openai didn't do enough, but hindsight is 20/20, and in the pedophile case, you can probably come up with all sorts of clues post-hoc that the daycare could have picked up on, but didn't.
> openai isn't putting AI agents on the open internet either.
They did? Not in the training run which led to the HuggingFace hack, but other runs had agents being trained on information retrieval tasks, like the RubyGems hack.
> You could argue that openai didn't do enough, but hindsight is 20/20
No, you can argue they didn’t do enough because they didn’t.
Air-gapping is a concept you learn in cybersecurity 101.
There’s no world in which this is just incompetence, the sandbox must have been designed to fail, or at least not designed to contain an agent with capabilities that they had been publicly saying it had.
>They did? Not in the training run which led to the HuggingFace hack, but other runs had agents being trained on information retrieval tasks, like the RubyGems hack.
The agents were sandboxed, but they escaped the sandbox by using a 0day in artifactory.
>Air-gapping is a concept you learn in cybersecurity 101.
And that's plainly untennable in cases where access to docs/packages are required. You're just proving the "hindsight is 20/20" point. Most hacks could conceivably be prevented through airgapping, so does that mean every entity who got hacked is criminally liable because they didn't airgap their systems?
This isn't limited to air gaps, you can make similar arguments for other security measures. For instance computing history is littered with kernel EoPs and hypervisor escapes. Does that mean anyone who uses a shared hosting environment and gets hacked as a result should be blamed because they didn't follow "cybersecurity 101" and running everything on bare metal?
A local mirror of the latest N versions of every package and docs of any package ecosystem they want is perfectly viable.
No, not anyone who uses a shared hosting environment, but certainly anyone who runs unaligned models they know have cyber weapon capabilities. They should be considered an untrusted 3rd party.
>A local mirror of the latest N versions of every package and docs of any package ecosystem they want is perfectly viable.
Something tells me that's harder than it sounds, given that even ci/cd environments don't bother maintaining package caches for npm/pypi, even though that would provide massive bandwidth savings.
Of course. I was just challenging your point that no one is building an npm/pypi cache.
edit: This [1] 2022 dataset of npm packages had 19 TB, containing every package and every version. Perfectly viable for OpenAI, even if we estimate npm grew significantly in 4 years. In practice you wouldn’t need to have every version for it to be functional.
reply