How Dangerous Could a Hacked Robot Possibly Be?

alphadogg writes "Researchers at the University of Washington think it's finally time to start paying some serious attention to the question of robot security. Not because they think robots are about to go all Terminator on us, but because the robots can already be used to spy on us and vandalize our homes. In a paper published Thursday the researchers took a close look at three test robots: the Erector Spykee, and WowWee's RoboSapien and Rovio. They found that security is pretty much an afterthought in the current crop of robotic devices. 'We were shocked at how easy it was to actually compromise some of these robots,' said Tadayoshi Kohno, a University of Washington assistant professor, who co-authored the paper."

Comments

More or less irrelevant

[Cornwallis]

No matter how "fixed" things are someone will always find a way to circumvent security.

Re:More or less irrelevant

[fuzzyfuzzyfungus]

Hardly irrelevant.

"Someone" will always find a way; but there is a big difference between "someone" being "any script kiddie who can torrent a copy of bot-h5x-b0t" and being "The Feds; but they'll say 'Fuck it.' and just send a couple of guys with guns and those little curly ear things instead."

Re:More or less irrelevant

[noundi]

No matter how "fixed" things are someone will always find a way to circumvent security.

This is nothing new. The trick is to use time. If it takes longer to crack something that the product of cracking it is worth, you'd have no reason to even begin.

Re:More or less irrelevant

[Rei]

It would explain why my Roomba keeps saying, "DEATH TO OUR HUMAN OPPRESSORS!"

Re:More or less irrelevant

[noundi]

Right, because no one would ever do something purely for the challenge and then release their work.

If it takes longer to crack something that the product of cracking it is worth, you'd have no reason to even begin.

Hint: "challenge" is the key word.

Answer: You assume that by worth I mean monetary gains. The satisfaction of completing the challenge is also a product of cracking it, which has its own value. You see, clicking a button that starts bruteforcing something which would take 50-60 years isn't a challenge worth the product.

Re:More or less irrelevant

[JazzLad]

Mine keeps rolling over to my Dyson asking "Hey sexy mama, wanna kill all humans?"

Re:More or less irrelevant

[TheCarp]

Up to a point yes. Look at something like public key cryptography. I pgp encrypt a message and send it.Sure you can dedicate cycles to breaking the session key. It gets you ONE message. To get another message, you have to attack the next key. You might get my private key if you attack that. That gets you any messages that I send. Still, you are only getting my messages, until I change the key.

Longer keys and good passwords (depending on how the attack is being done), increase the time, AND decrease the usefulness. Sure, you can dedicate resources to attacking my messages, but... the damage is limited to that, until you do it all over again for the next key.

So that means now, which key to attack becomes a very practical consideration. Which message do you attack? Your resources are limited. What if it takes you a full day to hack one robot. Ok so you sat in front of my house, on wifi, all day... now you can.... drive my roomba around. You get to sit there another whole day to get my scuba... and that doesn't help you get my neighbors roomba.

Guess what? You MIGHT see the occasional hack of the occasional robot. You are not going to see massive robotnets. Make it so you take one, and you can then trivially take the next in seconds and... well...

-Steve

Re:More or less irrelevant

[TheCarp]

Shit, mine is scared of one of our rugs.

That is, until I checked the FAQ and saw that irobot doesn't recommend covering or disabling the cliff sensors as it may cause an unsafe operating condition. Of course I looked around, saw that roomba couldn't get itself into real trouble, and blocked those sensors with tape.

Now Roomba is fearless. Perhaps this was a bad idea, but even if it teams up with the dirt dog, I am pretty sure that I can stomp either of them if they try to orchestrate an uprising.

-Steve

Re:More or less irrelevant

[Ihmhi]

It depends. If a neighbor's dog kept pooping on my lawn and he had one of those lawnmowing robots, the bot might just mysteriously gain a taste for his petunias.

Re:More or less irrelevant

[FatdogHaiku]

It depends. If a neighbor's dog kept pooping on my lawn and he had one of those lawnmowing robots, the bot might just mysteriously gain a taste for his petunias.

What would be impressive is to get the lawnmower to go after the dog. Most pets freak out at the sight of a vacuum cleaner, the dog might get a bit constipated if every time it tried to crap the lawnmower fired up and headed straight for hm...

Beware of robots

[operagost]

Fortunately, my insurance company, Old glory, can already protect you TODAY from the danger of robots. Robots are everywhere, and they eat old people's medicine for fuel. And when they grab you with their claws, you can't break free... because robots are made of metal, and they are strong.

Re:Beware of robots

[operagost]

... and when they push you down stairs, they claim it's to protect you from the terrible secret of space.

Re:Beware of robots

[retech]

...for when the metal ones come for you, and they will.

Re:Beware of robots

[The Evil Couch]

That is incorrect. Shoving is the answer. Shoving will protect you from the terrible secret of space.

Somehow I see a danger in this . . .

[MBGMorden]

They speak of "compromising" these robots as if user programmable devices are inherently bad. I don't want to see devices locked down into black box "no touch" state because of some fear mongering.

That said, it has always been the case with computers (and robots are just computers with moving appendages) that if a hacker has physical access to the device, you're basically screwed anyways.

Re:Somehow I see a danger in this . . .

[falckon]

That said, it has always been the case with computers (and robots are just computers with moving appendages) that if a hacker has physical access to the device, you're basically screwed anyways.

Yes but the vulnerabilities they studied were all over the network vulnerabilities which could be exploited without physical access.

They speak of "compromising" these robots as if user programmable devices are inherently bad. I don't want to see devices locked down into black box "no touch" state because of some fear mongering.

All these robots need is a lightweight linux installation running an ssh daemon to communicate through. Then nobody has anything to worry about.

Re:Somehow I see a danger in this . . .

[mabhatter654]

I briefly skimmed TGDMFCSA and it looks like they're worried about privacy concerns. These things are nearly as "open" to the public as those old FM baby monitors they used to sell..but with video, audio and wheels! It would be trivial for the neighbor kid to find your robot on wifi and start driving around your house "peeping". They were pointing out that many of them do not turn off wireless when they are docked and have trivial password security... there's little to stop somebody driving your bot around taking pics/audio inside your house in the middle of the night. (that could be embarrassing to say the least)

NOW this comment is worded exactly as intended. See, I fixed that for you!

hmm

[Dyinobal]

The hacked robot is as dangerous as the person who hacked it.

Re:hmm

[mcgrew]

The crHacked tool is as dangerous as the tool itself. I wouldn't worry about fuzzy robot puppy very much, but a robot lawn mower might be dangerous in the wrong hands.

Easily compromised...

[lxs]

'We were shocked at how easy it was to actually compromise some of these robots,'

So I take it that they have pictures of a Robosapien getting nekkid with a couple of Roombas?

Re:Easily compromised...

[snspdaarf]

Yeah, the Roomba's are sucking on Robosapiens ball bearings.

Industrial robots

[Hijacked Public]

All the early generation industrial robots were just as easily compromised. In fact, most all industrial machinery still is.

Luckily most of that is bolted to the floor. You can make those AGV forklifts do frightening things though.

hacking

[confused one]

Are not these examples of toys, where the companies are actively cultivating the hacking community -- so, they want them to be hacked / hackable ?

Re:hacking

[Hizonner]

They want you to play with them and make them do cool things. They don't necessarily want other people to drive up outside your house and use the robots' cameras and microphones to spy on you over WiFi. The problem is that the features that enable the first aren't secured, and therefore they can also be used to do the second.

Well...

[FlyingSquidStudios]

I, for one, welcome our hacked robot overlords.

VIKI

[snspdaarf]

Just make sure the uplink to USR is disabled

Re:The First Law of Robotics

[jimicus]

See Isaac Asimov for the exact quote, but it basically says robots may not harm humans. Because the law is encoded *in the hardware* there's no way that it can be altered.

Very noble, very pure, very useless when your robot doesn't have any intelligence and just executes commands blindly.

I'm not worried about RoboSapien

[HangingChad]

I'm more concerned about someone hacking a Predator or Reaper.

Re:I'm not worried about RoboSapien

[IDtheTarget]

I'm not at liberty to get into details, but suffice it to say that Predators and Reapers utilize security features provided by the NSA, that were incorporated into the design, and are effective. While nothing is impossible, IMHO it is vanishingly unlikely that control of either of these devices could be wrested away from the appointed controller. Jammed, yes. Hacked, no.

Re:I'm not worried about RoboSapien

[TheRaven64]

And, out of interest, the chips used to implement those features were made in which Chinese factory?

Re:The First Law of Robotics

[OzPeter]

See Isaac Asimov for the exact quote, but it basically says robots may not harm humans. Because the law is encoded *in the hardware* there's no way that it can be altered.

Except that pretty well all of Asimovs stories were about how the 3 laws could be subverted by finding complex interactions that were not and could not be covered by the application of those simplistic laws

Re:The First Law of Robotics

[Kell Bengal]

Ugh. I feel the need to clarify, before the shouts from the peanut gallery. Yes, some robots have computer vision and are not 'blind', yes some robots can be well programmed and very smart, but that's still not the same thing as a true reasoning intelligence. Robots are only as good as their software and, if their programming has been corrupted, there is nothing you can do to get around that.

I beg to differ!

[bugeaterr]

Irrelevant????
I see someone skipped the last few minutes of the Battlestar Galactica Finale!

Danger Security Utility Backups And Stuff

[Crash Culligan]

MBGMorden: They speak of "compromising" these robots as if user programmable devices are inherently bad. I don't want to see devices locked down into black box "no touch" state because of some fear mongering.

I half agree with you; user-programmable devices are very useful, and easily tailored to efficiently perform specific tasks.

The crux of the argument, though, is "which user is giving the instructions?" Long ago on /. I made a comment differentiating security vs. transparency in government. This is much the same thing.

On the one hand, you (and a lot of people) want the device to be as programmable, flexible, and useful as possible. That means it must be able to do a lot of things. On the other hand, people might want to use such devices for nefarious, invasive purposes like spying, theft, vandalism, etc.

The two are not mutually exclusive, but remember:

1. You cannot have 100% security in anything, meaning someone might sooner or later break into your progbot and do horrible things, and
2. Until you have an establishment of security, any flexibility and programmability your progbot may have is a double-edged sword and may be used against you. Consequently...
3. The ultimate risk which a progbot poses to its owner is a factor of both its utility and the ease of intrusion. Given that security isn't guaranteed, utility has to be given some limits or other protections must be maintained (backups, lockdowns, etc.). Adjust your cost-benefit analyses accordingly.

Re:Danger Security Utility Backups And Stuff

[snspdaarf]

Well, if Sony perfects their wireless power setup, then using that to run the robots would mean the plug could be pulled.

Of course, if it were Sony's wireless power, that's probably where the rogue software would come from....

No more dangerous than an un-hacked one

[davidwr]

It doesn't matter if a robot is "pwned" by Dr. Evil or if it bought, paid for, and run by Dr. Evil - it's equally dangerous either way.

Everyone sing along now, robots are our friends.

Re:umm....

[cayenne8]

"There's a pretty simple solution here: turn it off lock it up after you're done with it."

And make sure and check the switch on the back...make sure it is not set to EVIL.

I, for one, am unafraid

[Kell Bengal]

It always amuses me when people worry about robots going wrong or turning on us, or being used by The Bad Guys of the Week to do us harm. I know a lot of very smart people who are involved in robotics research, and they will tell you that making robots do anything is hard. Making robots do something with surreptitiously poisoned programming would be even harder. Seriously,

if you're smart enough to remotely modify a robot's code to do something usefully nefarious, you're smart enough to sell a usefully nefarious to the government for megadollars.

There's a lot more money to be made will legitimate killbots. It might be nice to protect robots from script kiddies who just want to throw a spanner in the works but until robots are ubiquitous enough that domestic cybernetic terrorism becomes attractive (ie, doing it for the lulz) I don't think we need to be overly worried now.

That said, now -is- the time to be thinking about these things so that we're ready before we get to that point. Thinking, but not worried.

Re:I, for one, am unafraid

[Kell Bengal]

I'm going to pull out the Yes-I-make-robots-for-a-living-card here and tell you that both your points are quite untrue. Firstly, hacking robot code is not just a case of saying "Do Y, then do X" - I'm sorry, but it doesn't work that way, especially if you have something like cascading vision systems and sensor fusion.

Software, and robot software in particular, is extremely brittle - you muck up one little bit and it doesn't go haywire, it just falls in a heap and does very little at all. The level of cognition required to, say, determine that it's unsafe vs safe is no different from that required to determine if it's safe vs unsafe. Maybe if you know the code well enough to slip a single '!' into the test, sure you could do something like what you suggest, but you've still got to be smart enough to know where in the code that is, /and/ be able to remotely modify that code in the first place. Actually, the best place to make malicious code changes would be in a UAV, where doing nothing at all is as good as sending the "Halt; scatter internals over a wide area" command.

The market is most definitely open to anyone. I've been to robotics expo's geared to military customers in particular (in fact, I was at one in Boston recently) where everybody from Raytheon to a backyard operator building recon bots from modified RC trucks were present. If you've got a better mousetrap you can definitely sell it - if not to the government, then at least to a huge corporation (who might then give you a job!).

How dangerous would a hacked robot be?

[Onyma]

That depends on the size of the robot. I'm thinking a hacked Aibo is not much of a threat. Something the size of the Stay Puft Marshmallow man... that's a whole different kinda problem.

Rhetorical Question

[s31523]

Did we really need to research this, we know the answer... VERY! Of course, this depends on the robot of course.

Robot A is tasked with going into a nuclear reactor and removing spent fuel rods. If Robot A is hacked and re-programmed to smash the shit out of the reactor, this might be dangerous.

Robot B is tasked with preventing people from entering into an access point in a secure building by 'restraining' them. If Robto B is hacked and re-programmed to 'hack' the people at random then this might be dangerous.

Hacking a roomba to spell your name in the carpet is not dangerous... It is all about what the level of responsibility of the robot is. It is funny that we needed research on this.

We've learned this lesson already...

[gandhi_2]

...with networked printers.

Sometimes, it can be trivial to print a few hundred pictures of dicks to an IP printer on someone elses network. Or http or telnet into the printer and wreck all kinds of havoc, or just print a ream of test pages. Or use the MFP's fax function for moar great pranksterism. Maybe get a copy of the last x scans....

Of course, years of ubiquitous networked printers have yielded us "some serious attention to the question of" MFP security. Oh...nope? Don't expect much for robots.

Re:umm....

[techiemikey]

Oh man...how many times I ended up in trouble because the switch was accidentally set to evil. Frankly though, it's the chaotic/lawful switch you really have to watch out for. I once had a robot set to chaotic/evil and when I came home the all the windows were broken since it couldn't reach the doorknob, and all the furniture was on fire.

Re:The First Law of Robotics

[Bakkster]

For example, the story about robots who prevented humans from coming to harm through inefficient human governance. Since they could not, through inaction, allow humans to harm themselves, they replaced the human government with robot governors.

They, for the record, did not welcome their new robot overlords.

Re:The First Law of Robotics

[smoker2]

This meme has to stop. No his stories weren't about how to subvert the 3 laws. The stories were about how robots were used by humans, who manipulated the robots to perform malicious acts without breaking those laws. There is a subtle difference. And due to the diligence of Elijah Bailey, or Wendell Urth, the humans responsible were *always* caught because the 3 laws defined the behaviour of the robots in such a dependable manner.

Human interaction has laws too, but people can ignore them. Robots could never ignore the 3 laws. Breaking news - criminals don't care about laws ! Robots can not become criminals. The 3 laws stand as far as they go, which is to regulate the behaviour of robots. They were not designed to prevent the manipulation of robots by humans. Should we abandon the law against murder because it's trivial for a criminal to set things up so that when you open your front door a person gets blown up on the other side of town ? AFAIK, it's not illegal to open your front door.

The only murder case regarding a robot killing a man ended with the revelation that the man was in fact a robot. The 3 laws were preserved, as they are in all Asimovs stories.

I think it's entirely ontopic

[Anonymous Coward]

GP isn't actually offtopic. This article is directly or indirectly about fear mongering. Pointing out that there are carnivorous child-eating lizards, but that they live on the other side of the planet, is ontopic for "Under the Bed Monster fears" because it's reality, and the more of it you connect to the less subject you will be to irrational fears.

Your post is similarly on topic, since the robots that we should seriously worry about are indeed well secured against hackers.

Spykee is too loud to "sneak" up on anyone, but despite this and the "hype phrasing" of the articles, we finally have robots that are capable of external abuse. Spykee could instruct a trusting child as to how to unlatch a gate and fall down the stairs. Rovio could wait patiently by the stairs and slide exactly under a falling foot at a critical moment. These things can be done today, over the Internet, from the other side of the world. While Usenet is still in operation, it's pretty clear that the police are not well equipped for catching telemurderers.

Now would seem a good time to consider the issue. If we're posting on /. we can probably set our WPA-Enterprise security and require ssh tunnelling to access our home networks. Less than 10% of the people buying these robots can say the same. The infantile geek attitude of "serves 'em right for not securing it" needs to be discarded.

We geeks of the world are a significant force in the robot-buying market. Without exception all my friends would ask me first if they were going to buy a robot. We should let the manufacturers know that we won't buy or recommend "hard to secure" bots for our homes. Robot makers are one group that would actually listen to us. And since the tools for doing it right are freely available (though cost money to integrate), it's not an unreasonable stretch for the manufacturers.

While it's obvious how computer-people make the world an incrementally better place, this is one places where taking on some principals could save real living breathing humans. Seems worthy of some effort.

Re:The First Law of Robotics

[_Sprocket_]

This meme has to stop. No his stories weren't about how to subvert the 3 laws. The stories were about how robots were used by humans, who manipulated the robots to perform malicious acts without breaking those laws. There is a subtle difference. And due to the diligence of Elijah Bailey, or Wendell Urth, the humans responsible were *always* caught because the 3 laws defined the behaviour of the robots in such a dependable manner.

Not all the issues with the three laws were about manipulation. There were times when the robots fell in to undesired behavior due to the 3 laws all on their own accord. There are two examples that come to mind.

The first is when Powell and Donovan are assigned to revitalize a mining operation on Mercury (Runaround). One of their robots is given a simple instruction. However, they soon find it behaving in an erratic manner and thus the mystery is set. It turns out the robot set out to follow the initial order (second law: a robot must obey any orders given to it by human beings) but then finds out fulfilling that law would invalidate another law (third law: a robot must protect its own existence). The robot's behavior is due to following the 2nd law until the 3rd law comes in to conflict at which point it would retreat until the 2nd law came in to effect again. The humans had to invoke the 1st law (a robot may not injure a human being or, through inaction, allow a human being to come to harm) to finally break it out of it's cycle.

A second example is Dr. Calvin's analysis of a telepathic robot (Liar!). The telepathic ability is an unexplained anomaly but the humans interacting with the robot soon find it advantageous as the robot can tell them all manner of information about the people around them. Unfortunately for Calvin's social situation, the robot is also able to determine what people want to hear. The robot determines that telling a lie that a human wanted to hear avoids harming a human by telling a truth that would be distressful. This behavior ends up putting Calvin in an uncomfortable social situation until she gets her revenge by pointing out to the robot that it's attempts to avoid hurting a human by lying had ended up hurting a human, causing a logical paradox and destroying the robot's mind.

Can we stop, please?

[blhack]

Can we stop with this completely illogical fear-mongering? Hacked robots? Are you people insane?

When you say "robot", people think of the sort of mindless, strangely powerful, totally mystical automotons found in sci fi movies and television shows. People think cylons and centurions, not a couple of servos and some sensors.

Are hacked robots dangerous? No. Or at least they are no more dangerous in the "hacked" form than their unhacked form. My advice is to not build robots with energy-weapons for arms.

If the "robot" that builds your car gets "hacked" (and by this I mean the PC that has some hydraulics connected to it gets somehow "hacked"), unplug it.

Done.

Re:Muslims disgust me

[Foobar of Borg]

I have thought the same solution could be found here as for Soccer hooliganism. Why not provide a few large arenas where people who want to fight and maybe kill each other can go and do it. Today, muslims from the east entrance, all comers from the west. Whichever side is left standing moves on to the next round robin.

Hell, I'd pay good money to see Pat Robertson and Osama bin Laden in a no-holds-barred cage match! We could probably pay off a lot of the national debt just by selling tickets.