... Why does a vacuum needs a webcam, a microphone, and Internet access? And what did the manufacturer do with all that data?
post
But he soon discovered that the same credentials that allowed him to see and control his own device also provided access to live camera feeds, microphone audio, maps, and status data from nearly 7,000 other vacuums across 24 countries. The backend security bug
I feel like "bug" is doing a looot of heavy lifting here.
It also illustrates how a security vulnerability is simply a bug, albeit a dangerous one.
Is it a bug though in this case? To me a bug is when a program behaves in a way that's not intended. This might very well be a case of the program behaving exactly as intended, except the intentions of the people who made it were wrong.
An online service is a program (or a bunch of program).
Giving access when it's not supposed to falls into behaving in a way that's not intended.
Therefore, an online service giving access when it's not supposed to can be classified as a program behaving in a way that's not intended.
Thus, this case fits into your very definition.
Giving access when it's not supposed too
Not sure I'd agree with that statement. Personally I see it as the correct credentials were provided, and thus access was granted; ergo, the app performed as intended, and there is no bug.
The error here seems to be around the lack of concern for security; nobody considered that using the same credentials for their fleet of robots could pose a threat if discovered. It's no different to someone using the same email and password for everything, and then wondering why their facebook has been hacked after their Reddit account leaked. The problem isn't a bug in code, it's just poor cybersecurity hygiene; what we see here is the same just on a commercial level.
[the robot vacuum] retails for around $2,000 and is roughly the size of a large terrier or a small fridge
Doing everyhing possible to avoid actual dimensions as always.
What size is a 'small fridge' anyway??
Long story short: he was trying to find the password for his own vacuum (yeah that already sounds ridiculous) so he could control it with a game controller, and found that the same exact credentials worked for an estimated 7000 other vacuums that need to call home to process visual data in the cloud. Hidden behind the lede: DJI automated vacuums require constantly sending their footage abroad to even work in the first place
Why do these things require microphones?
This honestly sucks
Good time to shamelessly plug valetudo, if your vacuum robot is supported.
With this, it does not access the public internet, and still functions the same as without rooting it. You just can't manage it if you're not home, unless you have some VPN set up or home assistant integration. But I don't know when I ever wanted to manage/watch my vacuum robot when I'm not home. Some sort of offline mode should be legally required for these kinds of devices that don't really need it. "Does not need an app to work" has become a major selling point for me for things, alongside "has physical buttons".
Also drop me a message if you're in switzerland and need an unsoldered valetudo breakout board, I still have around 5 left.
I see a new law coming, limiting the number of automatic bots/ai one person can legally give commands to.
The backend security bug effectively exposed an army of internet-connected robots that, in the wrong hands, could have turned into surveillance tools, all without their owners ever knowing.
Suprise, if it has cameras, microphones, gps, data mapping capabilities and speaks to a server outside of your home it is by definition is a surveillance tool.
All of this has always been predicated on the idea that the surveillance will never be used against you somehow. I don't know why anyone ever bought into that.
Which is why we need to de-normalize this thought process. People should always ask, why does my vacuum need to go online? Or at the very least, can I turn that "feature" off?
because vacuum makers lock away advanced capabilities on the app making a home network the only way you can access them.
of course, when you're shopping, it'll use those advanced features as advertisements; but there's no way to know that ahead of time unless you're buying the most expensive models available.
And to top that off, those features have a terms contract you need to accept to use. Yet you don't know what that entails until you spend a lot of money.
Both your point and this also needs to be denormalized. We should be able to have full visibility for purchase decisions and have no push back on returns if we don't.
if it has cameras, microphones, gps, data mapping capabilities and speaks to a server outside of your home
It has also happened with baby monitors. Well, almost endless other IoT devices too. But baby monitors are a particular issue. They have speakers in them too. Attackers were exploiting it and playing very disturbing sounds to cause extreme distress to babies.
I guess the most plausible explanation is incompetence, there wouldn't be a reason to do this on purpose (a backdoor), right? Since the company could have easily used different credentials per device that they store anyway?
Also this project for Neatos. https://github.com/Philip2809/neato-connected
Gosh, their drones are getting banned in the US. This just seems to help their case.
Long story short: he was trying to find the password for his own vacuum (yeah that already sounds ridiculous) so he could control it with a game controller, and found that the same exact credentials worked for an estimated 7000 other vacuums that need to call home to process visual data in the cloud. Hidden behind the lede: DJI automated vacuums require constantly sending their footage abroad to even work in the first place

all 6 comments