Nest defends web CCTV Cam amid unstoppable 24/7 surveillance fears Alphabet-owned Nest says there is no truth to the allegation that its internet-connected home CCTV cameras continue to record video even when switched off. This assertion comes after a report from ABI Research found that the Nest Cam keeps drawing a healthy amount of current even when told to turn off, suggesting it's still …
But SKY boxes are supposed to do stuff on standby, aren't they? On the other hand, are these cams supposed to be constantly listening for network commands to wake up? So maintaining a logged-in wifi client ... I guess that could use 0.3A. But, in that case, the CCD being on and (presumably) constant transmitting of video signal only uses an extra 10% --- is that realistic?
I'm interpreting Alphbet's carefully-worded statement to mean that the camera is fully powered up and recording to cache so that when it's switched on, there's data instantly available. The cloud connection also is still up... it's just not pumping data out to the cloud.
Key question I guess is: how big is the cache, and is the camera still recording to its internal cache when "powered off"?
That's a problem with pretty much every modern bit of kit, there isn't a true "off", it's basically always on and either in standby or full working mode. I can't even remember the last bit of kit I've seen that has a real, physical on/off switch, ie that breaks an electric circuit that hard-switches-off the device. Nowadays what seems to happen is you press a button which sends a software-level or firmware-level instruction to go to standby or shutdown.
We've gotten used to trusting device makers that standby and shutdown really do what they say, but its perfectly possible for all sorts of devices to be silently on when you think they're off. That's why recommended anti-spying practise with mobiles isn't to just turn them off but remove the battery.
"We've gotten used to trusting device makers that standby and shutdown really do what they say...
Nope. When I want one of my devices (okay, toys) to be off I'll make sure they are off. With some of them it's a bit of a hassle as they take some time to power up/boot, but I know that and can take it into account. Most of it isn't that time critical anyway, and a bit of applied time management can help, too.
I started doing this years ago when I started paying for my own leccy, and realized all the stuff in standby mode was adding up (and eating into the have-fun-at-weekends fund).
"That's why recommended anti-spying practise with mobiles isn't to just turn them off but remove the battery."
And that's exactly why you can't remove the battery from an iPhone.
As for the dodgy webcam, putting the damn thing in the cupboard should deal with it. Or even just chuck a cloth over it, parrot-cage style.
"It seems that rather than being a privacy menace that ogles you even when powered down, the Nest Cam is an energy hog that sucks up electricity even when you don't need it, in addition to being a privacy menace that ogles you even when powered down, and that this is an example of how Alphabet is maintaining the tradition of Google innovations that add value for everyone."
FFY.
Now then, anyone want to monitor the network activity of a Nest Cam when powered off?
What, like the authors of a scary "it's watching you 24/7!!!" report, who took the time to measure the current drain of the camera but couldn't be bothered to look at the network traffic to see if it really was transmitting video when switched off? It would have taken about 30s, not even requiring any equipment as specialised as the multimeter they seem to have managed to find in a drawer. But then they might not have got the result they needed to get the story slavishly reproduced by journos just ahead of Thanksgiving...
It doesn't mean much whether the the device is observed to be constantly "transmitting video when switched off." Maybe it transmits only a heartbeat. Maybe it records to some internal storage. Maybe it listens passively for a command, then starts transmitting video without lighting the LED. All we know is that the camera is active when it's supposed to be inactive. That's a real concern.
I would think a short review with Wireshark or other network monitor would tell the user to what extent it's contacting the mother ship when turned to off, (but not really).
My concern is to what extent remote monitoring and control is baked in.
Option: Pull the plug. Works every time.
"All we know is that the camera is active when it's supposed to be inactive. That's a real concern."
Why is that a real concern? If it is connected to the internet, which it needs to be to be turned back on via an app then even if it was almost fully powered down it could still be woken WOL style for nefarious purposes if required by a secret organisation. They'd just have to start it up about 30-60 seconds before they wanted to capture the action, not usually a problem in surveillance operations.
Other than that streaming video takes a fair bit of bandwidth in the upload direction. A rarely used direction for most households other than torrenters. You could easily use your ISPs bandwidth stats, a free netflow product (PRTG for instance) and see anytime it tries to upload and alert you.
If you really want to make sure it can't transmit then pull the plug. Every IP cam has exactly the same issue when plugged in.
If the power LED (turning from green to red to indicate 'off', instead of just plain turning off) draws 300mA, that cam would make a nice table lamp.
Indicator LEDs don't need to draw more than a couple of milliamps, 10mA if you want a nice bright one. Whatever the NEST cam is doing, it's not actually turning off.
"If the power LED (turning from green to red to indicate 'off', instead of just plain turning off) draws 300mA, that cam would make a nice table lamp."
Careful reading is required. Huns n Hoses didn't suggest that sort of draw was for the LED, but the difference in draw between powered up and supposedly powered down was possibly in the right ball park for the LED.
What they actually mean to say, is that instead of powering off like you tell it to, it goes into standby mode.
Off is exactly that. It's dead. No power, no activity, nothing. If it keeps drawing almost as much power as it's "on" state because it "has to wake up and start recording at any time", then it's in standby mode.
To label it otherwise, is false and misleading advertising.
Perhaps it is just sitting there, waiting to here someone mention a Google advertiser. When you mention Wal-Mart or something to your family, it kicks in and records your testimonial of your love for cheap microwave ovens and synthetic fabric clothes, and then stores that for posterity!!
That would be a very Google thing to do.
Presumably you can remotely turn the camera on from an app or web interface, and presumably you can do this from outside the WiFi network. In which case the camera needs to remain connected to a server where it can be told to turn on. That would require both the WiFi to be active and the CPU to be powered up.
I'd like to complain on behalf of the Lizard Mothership occupants and all other non human life-forms here on your Mostly Harmless world. We are getting a bad press - we don't do these things to you - we don't need to - you are perfectly capable of being your own worst enemies.
A camera module (for example the 5MP module on the Raspberry Pi) takes between 200 and 250mA when running. Seems a lot, but each pixel has 3 or 4 transistors, so that quite a lot of transistors to power.
Sensor take a fair amount of time to start up and produce decent images - let's say a couple of frames to get going, then 10 frames to produce a decent image (auto white balance, gain etc need to be worked out). At video rates that 300ms or more, the Pi camera takes 750m s to start producing decent images. Note that in the dark it takes longer as you might have to extend exposure times beyond the video rate (ie 30ms per frame) to get a decent image, so startup time can dramatically increase in low light conditions.
By running the camera all the time, you get round this startup time issue because the camera is always producing its best frames. You can of course ignore frames coming out of the camera when you don't need them.
H264 encode at 1080p30 takes about 70-100mA in a decent processor, usually more. Not sure if the Nest does h264 encode (might be mjpeg which needs a lot less power) , but the reduction in power quoted would seem to indicate that an H264 encoder would still be still running as it isn't dropping enough.
Of course, although you might have an H264 stream, you may not be sending it anywhere. Wireshark will prove that.
It's a Nest device.
When you tell it to shutdown it turns off the LED and cuts the network connection but the video stream is still active.
When you turn off the heating the sensors still measure temperature.
When you turn off the burglar alarm it still monitors the movement sensors.
The Fire/Smoke alarms are 'always on' but only trigger on 'event'
When one device triggers (either time based or event based) the network connection is re-established, all devices in the (local) Nest awaken and empty their buffers to the cloud.
The triggered device will process the 'event' and other devices resume their pre-event status.
People who buy into this Nest/Hive/Swarm/Murmuration stuff are quite happy with the tradeoff same as FB users.
Nobody forces them to give up their data, they do so willingly.
"...But, in that case, the CCD being on and (presumably) constant transmitting of video signal only uses an extra 10% --- is that realistic?"
Yeah. Wifi chips are rather power-hungry, and the (I assume ARM) CPU is probably using some power. CMOS camera chips (it's probably not CCD) use in the order of 10s of miliwatts. W=V*A (watts = volts * amps) so this means under 10milliamp curent even if the Nest is 3 volt.
"What they actually mean to say, is that instead of powering off like you tell it to, it goes into standby mode."
Given the high power use, I doubt it's even going into a true standby mode (i.e. putting CPU or wifi chip into a lower-power but slower-to-respond mode.) Needless to say, I figure if you go to the trouble of turning something like this "off" it should get MUCH closer to "off" than this -- it's no problem if it takes a few seconds to get ready when turned on. (Satellite TV boxes get a pass on this to some extent, since they are potentially recording shows when "off"... but a device like this that should actually be doing nothing when "off" has no reason to use that much power.)
As the Nest cam can be remotely upgraded and has to listen for "start recording" events, I'd fully expect the CPU and WiFi to remain powered up. Imagine turning the camera "off" and then discovering that you can't record the burglary in progress that you have been notified about by some other means, because the camera is off and not just in standby.
And as for remote automatic firmware upgrades - you obviously don't want those happening when the camera is powered up, connects to Google and starts downloading and installing new firmware when it is supposed to be recording video.
It's fixed to be turned on remotely without notice.
So many devices are on the market which have obvious secretive mass surveillance capability baked in I must wonder if secret or not so secret laws are in play mandating certain panopticon functions and allowing manufacturers to lie profusely about it if caught.
Like maybe this time.
or accept the fact that if it is connected, then there is a chance you could be spied on, as with everything else connected to the Internet. Most of us however are too insignificant to make any illegal activities directed at us profitable, and we have nothing worth hiding anyway.
Whenever I service laptops for family members,I always tell them that their cam should always be assumed to be turned ON.
Knowing that,it becomes easier to convince them of the most effective way to ensure their privacy from hacks.
I stick some black insulating tape over the lens on laptops,if they are silly enough to want to talk over the net whilst displaying their funny bits',well, that's up to them & they are also likely to 'forget' to cover it up afterwards.
I have never ever used a web cam for it's intended purpose,I have used a CCD type for astronomy,but I have never ever 'trusted' them or the software.
Considering ALL WINDOWS machines are inherently vulnerable,thanks to Bill GATES,I think it's wiser to just use your mobile phone to communicate,even though HMG snaffel every bit of data for their egocentric paranoia.
Sounds like an excuse to me, maybe they didn't "mean" it ?.. Can be proven one way or another just by hanging a scope on the video signal output and wave hand in front of the camera. That or use debug tools to see what's happening at process and network level. Personally, I wouldn't trust them an inch without verification. Nullius in verba and all that..
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
