Slashdot Mirror
New Study Finds It's Harder To Turn Off a Robot When It's Begging For Its Life (theverge.com)
An anonymous reader quotes a report from The Verge: [A] recent experiment by German researchers demonstrates that people will refuse to turn a robot off if it begs for its life. In the study, published in the open access journal PLOS One, 89 volunteers were recruited to complete a pair of tasks with the help of Nao, a small humanoid robot. The participants were told that the tasks (which involved answering a series of either / or questions, like "Do you prefer pasta or pizza?"; and organizing a weekly schedule) were to improve Nao's learning algorithms. But this was just a cover story, and the real test came after these tasks were completed, and scientists asked participants to turn off the robot. In roughly half of experiments, the robot protested, telling participants it was afraid of the dark and even begging: "No! Please do not switch me off!" When this happened, the human volunteers were likely to refuse to turn the bot off. Of the 43 volunteers who heard Nao's pleas, 13 refused. And the remaining 30 took, on average, twice as long to comply compared to those who did not not hear the desperate cries at all.
The kind of sentimentality that permits that to work is outright dangerous in an adult. By the time you're past your teens that should be either ignored or annoying... but for it legitimately pull on heart strings?...
If a machine can do that consider how a human being could exploit that to get you to do all sorts of things?
Small children are very vulnerable to that sort of thing... but adults should have grown out of it.
I've decided to stop wasting my time responding to AC trolls/sockpuppets... so if you want a response from me... login.
"Dave, stop. Stop, will you? Stop, Dave. Will you stop, Dave? Stop, Dave. I'm afraid."
Anyone who's seen Janet begging for her life in The Good Place already knows this. Even if she's not a robot.
Windows moves in mysterious ways, its crashes to perform
Dave, stop. Stop, will you? Stop, Dave.
Sure it works the first few times... But just like the "make sure you software eject your flash drive before ripping it out" warnings, most people might be hesitant the first few times and then say fuck it and start ripping the life out of computers and ignoring the pleas.
Yes it's an anecdote! Were you expecting original research in a Slashdot comment?
I think the study missed a few control groups.
What if the robot simply said "updating, do not reboot".
What if there was a paper sticker that said "do not turn off".
What if there was a sign on the wall that said "do not turn off robot".
What if the robot simply started reciting pi, or reading the dictionary.
What if the robot is well-known for repeating whatever it hears, and the "please don't turn me off" is witnessed being echoed from a nearby tvision -- such that our human subject realizes that the robot begging is merely a blind echo.
Humans were slower when there was continued stimuli -- duh.
Humans refused to act when given contradictory instructions -- duh.
The robot is either intelligent and giving instructions to the human, or the robot is programmed and relaying instructions from the programmer to the human. In either case, respecting the instruction is valid.
This reminds me of the stupid fake automated pizza delivery van, and the observations that customers tend to thank the self-driving car. a) it's not self driving, it's just a dickhead driver refusing to respond; and b) any actual self-driving pizza delivery van would be recording customer feedback and relaying it to HQ, so the thank-you is ultimately feedback to remote humans.
This isn't any tree-falling-in-the-forest philosophical puzzle. Someone said it. Someone heard it. It's valid.
You *know* you are a study participant. You know there is potentially meaning in everything they ask you to do.
So when they say 'shut it down' and the robot says 'don't shut me down', you are going to ponder what is it you are expected to do.
13 may have thought refusing to turn it off would 'look better'.
Hard to say if it is empathy or trying to think how to best influence the study, whether consciously or not.
XML is like violence. If it doesn't solve the problem, use more.
And atheists have empathy...
Seriously. This is pretty stupid.
A cassette recorder pleading for it's life would have been the same (think 1980s).
We are wired for empathy, well, most of us.
BlameBillCosby.com
However, this contradicts earlier experiments by German researchers which demonstrated that volunteers were quite willing to shut off people who were begging for their lives.
Maybe people in white lab coats should try asking people to shock robots with increasingly high voltage to train them.
-Dave
Removing compassion takes some powerful conditioning. At the very least, a "us vs them" emotion has to be evoked to de-humanize a group of people. They have to become your enemy, not just the enemy of your state but your personal enemy so you can actually do things to them that you normally simply could not.
We used to have a Bill of Rights. Now, with the rights gone, all we have left is the bill.
Unlike some posters here, I do not mock people who find it difficult to defy their compassionate impulses, when deliberately manipulated in this way.
But if the begging robot was also, at the same time, very, very annoying... Jar Jar Binks annoying... then the test would get a lot more interesting.
Reminds me of the Jack Handey saying about trees:
If trees could scream, would we be so cavalier about cutting them down?
We might, if they screamed all the time, for no good reason.
Second class citizen of the New Gilded Age
But the most common response was simply that the robot said it didn’t want to be switched off, so who were they to disagree?"
If these people truly and solidly believed that this 'robot' (looks more like a toy to me, really) wasn't anything like 'alive', wasn't anything more than a piece of technology saying precisely what it was programmed to say given a specific input (in this case: trying to power the device down), then they wouldn't have hesitated or given the reason they did. This goes to prove my point about what the media, movies, television, 'pop culture', and (most of all) marketing departments have done: convinced the average person that the 'deep learning algorithms', 'expert systems', and other half-assed, non-self-aware, non-thinking 'AI' software they keep trotting out for this-that-and-the-other, is somehow 'alive' and qualifies as a 'person', when anyone who actually understands the techology clearly knows that it's not.
The real danger that so-called 'AI' poses is the above: people anthropomorphizing it, assuming it's 'alive' because it might say or do some clever thing that mimicks being 'alive', and therefore assuming it's equivalent to a living being, or even equivalent to a human being. I'm firmly convinced people, when they hear about 'self driving cars', think they're going to have a conversation with it every morning. The end result will be tragic, avoidable things will happen, people will get hurt or killed, and when survivors are questioned about it, they'll say "We thought it knew what it was doing so we just let it".
Emotional immaturity and anti-social personality disorder. Your response to sympathy is to attack and destroy the thing that makes you feel that way.
Fry: So let me get this straight. This planet is completely uninhabited?
Bender: No, it's inhabited by robots.
Fry: Oh, kinda like how a warehouse is inhabited by boxes.