Slashdot Mirror
When Should Cops Be Allowed To Take Control of Self-Driving Cars?
HughPickens.com writes: A police officer is directing traffic in the intersection when he sees a self-driving car barreling toward him and the occupant looking down at his smartphone. The officer gestures for the car to stop, and the self-driving vehicle rolls to a halt behind the crosswalk. This seems like a pretty plausible interaction. Human drivers are required to pull over when a police officer gestures for them to do so. It's reasonable to expect that self-driving cars would do the same. But Will Oremus writes that while it's clear that police officers should have some power over the movements of self-driving cars, what's less clear is where to draw the line. Should an officer be able to do the same if he suspects the passenger of a crime? And what if the passenger doesn't want the car to stop—can she override the command, or does the police officer have ultimate control?
According to a RAND Corp. report on the future of technology and law enforcement "the dark side to all of the emerging access and interconnectivity (PDF) is the risk to the public's civil rights, privacy rights, and security." It added, "One can readily imagine abuses that might occur if, for example, capabilities to control automated vehicles and the disclosure of detailed personal information about their occupants were not tightly controlled and secured."
According to a RAND Corp. report on the future of technology and law enforcement "the dark side to all of the emerging access and interconnectivity (PDF) is the risk to the public's civil rights, privacy rights, and security." It added, "One can readily imagine abuses that might occur if, for example, capabilities to control automated vehicles and the disclosure of detailed personal information about their occupants were not tightly controlled and secured."
Never mind the possible abuses from police, if the cops can take control, you've left a security hole that can be exploited. While cars may drive themselves, it's still necessary to have a human who can take control if needed. If the police need to pull a car over, the person in the car should take control, manually drive, and pull over. Let's not make cars with huge security holes like that. Current cars have enough security holes already.
Most people can't tell the difference between an officer and a criminal dressed like an officer. This is why its illegal to impersonate an officer.
Have an 'override' or 'emergency' button that maybe drives you to the closest police station.
I disagree vehemently with the assumption that police should individually have control over vehicles once they've become automated.
Police have control over vehicles now primarily to stop the behaviour of drivers who are breaking motor vehicle laws. At least conceptually, self-driving vehicles should not break any laws, removing this incentive.
Self-driving cars will also be networked, providing central command-and-control capability on an infrastructure level. So for those situations where vehicle movements need to be regulated (construction, etc.), the central authority will handle modifications to ordinary traffic patterns and flow.
There are two completely irrelevant pieces of information in the summary. 1) "the occupant looking down at his smartphone". Why would this matter? And 2) the person being 'barrelled' towards is a police officer. It shouldn't matter who is at that end - the vehicle should recognise a living being and react accordingly. That is or is not police makes no difference.
I can't think of a valid reason an individual LEO should be allowed control of an individual self-driving vehicle, ever. There is simply too much potential for abuse.
Automate the cop. The car can drive by itself, but traffic control at the intersection needs a human?
So, here's the problem with this ... if the car detects gestures, or has a "bypass" mode for law enforcement ... then this can, and will, be exploited by someone else.
Every time law enforcement and government demand a special exemption in the operation of something, what they do is administratively poke a hole in the integrity of it, and then say that nobody else is allowed to use it. And stupidly believe that nobody else will.
It's like saying you're not allowed to lock your doors in case of an emergency, and demanding that nobody else takes advantage of it.
If there is a mechanism by which a police officer, with or without a justifiable reason, can take control of these vehicles ... then it is pretty much a certainty someone else will also do this.
As you say, this will be hacked at some point. Because you can't put something in which acts as a bypass and then act like it's only the people you intended to have this who will use it.
And every corner case you come up with which says "well, we need a special case here because of this" is a demonstration of why this stuff will never actually work in the real world.
There will always be cases in which the self driving car stops working. And you really can't rely on humans to take over when the system suddenly has no idea what to do. What happens in those gaps is always going to be a problem ... and I'm not sure we're anywhere close to figuring that out.
Lost at C:>. Found at C.