If malware via monitor cables is a matter of national security, this might be the gadget for you

GCHQ's cyber arm has entered the hardware game with its first device designed to prevent cyberattacks on display devices. Called SilentGlass, the small gadget's intellectual property is courtesy of the UK's National Cyber Security Centre (NCSC), and the signals intelligence agency licensed it out to UK-based Goldilock Labs to …

[wolfetone]

And I am sure the NCSC wouldn't have baked in the ability for them to have a nose at what's going through it either I expect?

[RockBurner]

Well - if the device is going to remain in any way relevant, then it's firmware would (presumably) need to be updateable. There's your access point for telemetry to the mothership, backdoor snooping, and potential uber-hacker access point all in one. :D

[Irongut]

Who says it has firmware? Could be an RF choke in a box for all we know.

[Tron]

Could this be the UK's answer to AI?

A pointless product that does nothing bar snoop on your data, and you have to pay for it.

[Anonymous Coward]

> Could be an RF choke in a box

That is *exactly* what I came here to say! That's my bet as well.

(With blinkenlights and all)

[CR]

Also pretty expensive, because magic.

[ITMA]

We need to get one to BigClive to disassemble:

https://www.youtube.com/@bigclivedotcom

Just to make sure it isn't like one of those "un-trippable breakers":

https://www.youtube.com/watch?v=2TJEzdqtXlQ

[Anonymous Coward]

Of course not! VGA never had this problem?

[Anonymous Coward]

Any device like this with an association to GCHQ is more likely to be spying on your video output than protecting you from attackers.

[Jou (Mxyzptlk)]

4k120 and VRR?

Sounds overkill for office, but 120 Hz is nice to work with. Moving windows is so fluent, and it feels faster on click reaction (exception: Windows 11 UI). And VRR just "'cause duuuude! Of course VRR!"

(Oh, VRR = Variable Refresh Rate, nvidia G-Sync for example)

[andy the pessimist]

coax?

The HDMI maximum power is about 291mW. That will radiate. How far , possibly a mile. If people have a good aerial possibly more. The aerial is not discrete.

A coaxial shell would be an answer.

A metal mesh in front of the monitor screen would help.

All of this require close proximity to the building. Civil servants look out the windows.

[Kurgan]

Re: coax?

HDMI cables are already shielded. At least the good ones are,but probably monitors are leaky, too, and not very well shielded. About this whole idea, what about HDCP? HDMI data should be encrypted by HDCP, to stop us pesky pirates from copying DRM encumbered shit (it seems it's not working as intended, but I digress). So HDMI leaking EM signals should be encrypted, too, I suppose. Picture those pesky Chinese government hackers being thwarted by Hollywood's DRM scheme, LOL.

[andy the pessimist]

Re: coax?

I was assuming it was signal capturing. For malware in a secure area i would expect IT or BOFH to keep the network clean.

[Phil O'Sophical]

Re: coax?

about 291mW. That will radiate. How far , possibly a mile.

Voyager 1 has 100x that power, and manages 15bn km, it all depends on the antennae...

[Paul Kinsler]

Re: Voyager 1 has 100x that power, and manages 15bn km, it all depends on the antennae...

... and the level of background noise, I would imagine.

[Giles C]

I read it differently.

What I was looking at was some form of malware that could be uploaded to a monitor through the interface to infect another machine when it is connected to the screen. Considering that hdmi and usbc both support networking natively, it seems that that could be an overlooked vector. Go to a coworking space or similar and connect to a monitor, come home with a nice piece of malware.

[Kurgan]

I'm thinking more of a monitor that has been made specifically for spying, not one that has been pwned. I'd say that a "normal" monitor does not have enough capabilities (memory, etc) to contain a malware injection tool. But what about a malicious monitor, built for the job? It could work, exactly as malicious USB power supplies.

But if we come to this, what about printers, mice, keyboards, and every other usb device that is not a battery charger? The need to communicate, you cannot simply cut them off. To sum it up, I think that a monitor is quite an unusual attack vector. But maybe it's a good one exactly because it's unusual. Nobody expects the Spanish Inquisition.

[IGotOut]

100% monitor to pc. Monitors have all kinds of connectivity, so given the crap you can do with a USB device (ID spoofing so the host thinks it's a keyboard) , it's possible that IT have locked down the USB ports on the PC, but this becomes another way in.

[DS999]

The malware could be used to infect the SAME machine

The privilege level needed to compromise the monitor (or just send it a firmware update if security around that is rather light) may be less than the privilege level the system is operating at when it does EDID etc. to identify the monitor.

So if you can hack it from a userlevel process, then have the monitor p0wn the system at a root/Administrator level the next time the system is booted or another redoes the EDID process then you've elevated your privilege level even if the system itself was (somehow) totally secure against root level escalation attacks.

While ethernet over HDMI exists there is little support for it - and AFAIK it requires a special HDMI cable to support it so it is probably not a practical attack except in very very limited circumstances.

[Irongut]

> The Register had a bunch of questions, but the NCSC refused to answer any.

Until NCSC tell us what this thing is, how it works and what it protects against it lies somewhere between snake oil and spyware.

[Kurgan]

Exactly. Suppose these devices contain a cellular modem and they are actually made to spy on your hdmi data going through.

[jdiebdhidbsusbvwbsidnsoskebid]

If they are for sale, I fully expect some security researchers to be doing a tear down very soon.

I wonder what ifixit's repairability score will be?

[doublelayer]

The sales promises add credence to this possibility. They can't tell us what can actually be done over an HDMI cable, but they can promise that their device blocks all the threats. Given the complexity of HDMI as an interface, I can believe that there are problems in HDMI stacks which could be exploited, and if you knew about them, then a device that looks for them and blocks them would guard against those threats although one that looks for them and sounds an alarm would be more useful. But that wouldn't be threat-agnostic unless it simply blocked some channels, and those channels presumably have a point or you could block them yourself. Maybe it's as simple as disconnecting some things that they assume people buying this, who are probably using monitors in an office environment, probably aren't using, assuming that those pathways could be abused somehow.

[KarMann]

But it also protects against tigers and ice giants! Do you want your (hypothetical) children to be eaten by tigers and ice giants?

[Anonymous Coward]

A late entry for April Fools Day?

[Paratiritis]

New target

So now bad actors will have to use a Mythos type of setup to discover any vulnerabilities in SilentGlass.

[smudge]

Goldilocks - just right, or too good to be true?

these devices are equipped with hardware that identifies malicious traffic in the data channel, blocking the transfer between computer and display.

I'm thinking that it's a bit too late by then....

We're also told that the SilentGlass gizmos are threat-agnostic, meaning they are capable of detecting any kind of nastiness

Checks calendar - nope, April 1st was a while ago.

Ignores obvious comment about filtering out anything from/about the US Government.

Wonders how they can detect any kind of nastiness - including those not yet invented. And how it can let through all the stuff that you want, without labelling any of it "nasty".

Thinks we must be getting into Godel/Turing incompleteness/undecidability territory there.

Then remembers that Goldilocks is a fairy tale.

[I could be a dog really]

Re: Goldilocks - just right, or too good to be true?

I read it as it allows known valid traffic and blocks everything else. The everything else is easy as there are only a small number of valid operations. It sounds like monitors may be "lax" in processing and can be subverted by invalid or badly formed messages - and I can beleive that.

[JWLong]

What is This?

Another solution looking for a problem!?.....

[Anonymous Coward]

Device blocks malicious traffic between computer and display :o

> .. we are reliably informed .. these devices are equipped with hardware that identifies malicious traffic in the data channel, blocking the transfer between computer and display.

So, it sends all your traffic back to GCHQ.

[EnviableOne]

My 2 penneth

The use case is probably using the monitor as an Exfil channel, similar to all those let's use fans or LEDs, or something else that exists, to send data to a remote receiver from an air gapped network

It will be a kill anything outside of the visible spectrum type filter

[Frank Bitterlich]

I have doubts...

So there is a new device that claims to protect us from hypothetical threats, all of them, regardless of the type of threat or which method they use, everything "malicious" is being filtered out, without hampering the the actual use of the data channel for legitimate purposes, and they can't tell us how it works, we should just trust them.

Is it me, or does that sound totally crazy?

[firehorse]

Re: I have doubts...

Doesn't sound that crazy to me. HDMI is a well defined standard - and I would assume any legitimate data passing over the data channel adheres to a fairly limited set of types/structures/content - and it would not be beyond the realms of possibility to flag anything that clearly did not conform to the HDMI standards.

[Frank Bitterlich]

Re: I have doubts...

I guess that may be true for the Display Data channel, but I'd expect any meaningful hacking happening over the Ethernet channel.

But even validating the DDC traffic would be a tall order for such a tiny device, considering the hodgepodge of different protocols potentially running over that channel... not impossible, but a pretty ambitious goal. I'd rather expect it to break some more exotic (but legitimate) uses.

[Anonymous Coward]

I suspect this is more a "one way valve" to prevent data being fed back to a host machine but there have been proofs of concept using HDMI chips as software defined radios so...

[Pen-y-gors]

Cheap alternative

At least for the problem of the cable leaking signals.

Go down scrap metal yard

Buy 6 foot of old lead water pipe

Run cable through pipe.

Upgrade is to nick the lead sheet off the roof of your local church, and cover the floor/walls/ceiling/windows/doors of your room. May interfere with mobile reception.