What’s a logical reason to give a war robot emotions?

Question

During a bloody and violent war, the government orders a team of 1,000 scientists from many different fields to create a robotic machine capable of turning the tide in the war. The design an android, called Alpha-x167980 or Death Machine.

It’s outer shell is made out of titanium, and it’s delicate software is incased inside bulletproof Kevlar. It is powered by a cold fusion nuclear reactor. It is ten times as strong as the strongest human, it has a reaction time twice faster than a human, can run at 30 mph and knows how to use nearly every weapon and martial arts.

It has a modular arm that it can equip a machine gun, flame thrower, or mini gun and It’s other fists has a retractable blade.It has night vision, and can see in infrared and ultraviolet spectrums, and has heat vision. It also has rocket boosters that can be used to give it a boost in combat. And it is effectively immortal, as it can go to a copy of its body if it is destroyed. My question is, since it’s basically a walking tank with enough weaponry to overthrow a small country, why would they program it with emotions?

EDIT: In my question, I mean to say is their a logical reason to make a robot self aware and able to gain a personality form experiences.

Answer 1

The emotional warbot is a more agile, more creative, and overall more efficient warbot than a purely rational one.

And that's because of what emotions are, when you get down to it. An emotion is a cognitive shortcut distilling large amounts of information into a simple feeling that can be rapidly processed and actioned.

This has clear advantages- a warbot that has a bad feeling and avoids a rocket by suddenly hitting its thrusters is more effective than the one that takes a split second longer to analyse the situation rationally and decide to move (by which time it's thoroughly exploded).

This also means that logical trains like 'this officer has a proven record of good decisions therefore I will weight their orders 25% higher than the other equally ranking officer...' get distilled to 'I like this guy.'

That this occasionally leads to one of the warbots serenading its operator, promising to sweep him away onto its T345 Prometheum Flamethrower and carry him away to a special place just for the two of them, once the war is all over, is just a happy accident.

Answer 2

To make it humane.

If some general would release this on an enemy camp it has to be able to have mercy, to understand morality, so it could deal with enemies, but it would not slaughter the children, just because somebody gave them a gun.

Also understanding of human emotions could help it distinguish a real order from something said out of rage or meant as a joke.

Finally, to make the robot more "intelligent" the scientists used "neutral patterns"(or other technobabble) of human brain. The side effect was made into a feature, or perhaps that is how they developed the emotions for the robot.

Answer 3

It was not always a war robot.

I once was in a bar and noticed my seat had seatbelts, as did many of the other seats around. There was no prospect this bar was going anywhere. Why would there be seatbelts? The seats had been salvaged from a plane.

So too your warbot. The makers were under tremendous time pressure. The basic software from the robot was hurriedly adapted from available software platforms, copying heavily from teacherbot and copbot programs. Once the basic functionalities had been achieved these things were turned out as quickly as they could be produced, and they functioned well. But elements of the old programs remain.

This could be a phenomenal short science fiction. War humans were also something else before they were war humans.

Answer 4

To know its enemy.

I don't think anyone has said it yet, but the obvious reason is for the robot to better understand its human targets. If the robot is going to go into a situation to either cause maximum terror to the enemy or to win the hearts and minds of the local civilian populations it needs to understand, in the given situation, how a human would respond emotionally. To have the empathy required for these situations, the robot needs to be able to feel the emotions of humans.

This allows it to not only better predict the next actions of its emotional human enemy, but it can also attempt to drive them into reckless actions by influencing those emotions.

It will also be more accepted by civilians if it's more than an unfeeling war machine.

Answer 5

They wouldn't. It would be like making your toaster gluten intolerant.

The whole point of training soldiers is to remove empathy, compassion and other human emotions so the soldiers do as ordered.

Humans can be inhumane. ISIS soldiers will happily strap a bomb to a child and send the child off to kill some infidels. American soldiers have to shoot said child or get blown up.

Giving emotions to a killing machine serves no purpose in war as it could lead to ambushes such as a wounded child hiding a bomb or as bait into a trap.

Now to give said robots emotions, there are three reasons I can think of.

First, one of the creators feels guilty so hides code in the OS with a trigger word to give emotions should the robots be used to oppress the people.

Secondly as a fail safe should something go wrong and the AI decides to go all terminator on humanity. Removing the code to obey orders automatically releases empathy and emotions.

Thirdly the enemy develop a hack or virus to give emotions to stop the robots and prevent more from being made.

Answer 6

The short answer to your question is that you wouldn't. It would be a really (REALLY) bad idea (and dangerous to boot) if it's even possible, which I don't think it is.

Let's start with a bit of simple neuroscience. Over the course of evolutionary history, brains have developed 3 primary functional centres; the cerebellum, the Limbic System, and the Cerebral Cortex.

The Cerebellum (hard wired electrical) is the seat of autonomic functions, like keeping the heart beating, etc. It's also the source of primary instincts, like survival, hunger, procreation, etc. These instincts are designed to drive the animal in question, force them to do certain things to ensure the survival of the animal itself and its species. Trouble is, having hard wired responses don't always lead to the right result, so a new area of the brain evolved, designed to override instincts under certain conditions.

The Limbic (chemical) system, also called the 'Reptilian Brain' is the seat of emotions. In certain contextual situations, it's important for behaviour to be driven in a manner that overrides instinct, like protection of young at personal risk. It also stimulates hormone production, like adrenaline, to encourage best possible action in a fight or flight response moment. This is also a motivating area of the brain, but it too can cause you to make the wrong choices in certain contexts, and doesn't cater for situational (contextual) responses, hence the evolution of the Cerebrum.

The Cerebral Cortex (soft wired electrical) is the 'programmable' part of the human brain and allows for learnings to be developed within a single lifetime, rather than an extended period of evolutionary development. Also known as the mammalian brain, it is the seat of reason and what we consider intellectual learning.

(All this is a simplification, but functionally accurate)

Computers ONLY replicate the cerebral cortex functions, and even then only to a very limited degree.

The point of this is that computers DON'T experience a survival instinct, don't get happy or sad, or angry, and wouldn't understand what those emotions were even if they did. Skynet simply can't happen; at least, even if a computer system could become aware, it doesn't follow that it would act to preserve itself in any way. It simply doesn't have that innate motivation.

Sure; you could program it to prefer survival, but if you really want a machine to start trying to wipe humanity out, you don't need an AI to do that. I could program a drone to fly around, shooting anything that moves, returning to an automated refueling and rearming base whenever it's low on supplies, and that would take little more sophistication than a sophisticated modern video game.

Additionally, because emotions and instincts are NOT programmable, any attempt to simulate them in computer code would only be exactly that; simulations. The computer would not actually experience these emotions, but would emulate that experience, which in turn would only serve to confuse humans interacting with it.

And there's your only real answer; if you want the machine to be some sort of infiltration device, confuse the enemy into thinking that it can be reasoned with, then you introduce emotions so that humans begin to relate to it and anthropomorphise it. But, that robot won't actually feel emotions regardless. In that sense, the introduction of emotional responses would be for human benefit, not robotic.

Answer 7

To make it a better killing machine.

Many people are talking about how the machine should be more humane, and how its impulses to destroy should be somehow balanced by compassion and understanding.

I say that you don't have to give it all human emotions.

A machine that is programmed to point a pipe at moving targets and shoot will be mathematically and robotically efficient, yes.

But an AI programmed to feel nothing but rage and revenge will be efficient and motivated. It could be John Wick motivated. Give it a broken Aibo and convince it that your enemies destroyed the puppy, then watch from a safe distance.

Answer 8

The single hardest problem in all of AI today is developing a language to describe precisely what you want the AI to do. It's relatively easy to write languages that work for short term goals like "kill the guy in front of you wearing a red shirt." But the longer term goals humans have are hard to describe in a meaningful language.

Having the AIs learn during peace time would be the hardest. Phrasings like "Be ready for the next war" run risks akin to the paperclip-maker arguments: the maximum best preparation for war may involve consuming valuable resources at a disastrous rate. More precise instructions which let the AIs train themselves for months on end are terribly hard to phrase formally.

We know of one way to be able to understand all of human language: be self-aware. Many of the concepts which make our language useful are built around self-awareness. It may be the only way to communicate our deepest desires.

Which, to me, makes for an interesting paradox. We are always worried about skynet, when it comes to robots. Making it self-aware is always the problem. But in this sense, it is also the solution. The only way for the robot to truly understand the human condition is to be self-aware, and there's a reasonable argument that that means we could communicate with it and reason with it.

In this sense, we should not be afraid of self-aware robots. We should be afraid of the robot that was self-aware, and lost it. It's not when skynet becomes self aware that the problem starts... it's when skynet loses it.

Answer 9

No,there's no logical reason. But that's not the whole answer.

Your robot will be judged by its functioning. Whether we call its decision making method "logic" or "emotion", we will want it to make certain decisions, as we intend, and not make certain others. We will want predictability. That includes wanting predictable unpredictability - we expect it to move randomly under fire to evade harm, and we will test various random and quasi-random methods till we find "the best". If we want it not to kill certain people or to do certain things, how it decides that is completely irrelevant, we will pick whatever method we find most effective. Specific directives in some cases, machine/neural learning in others , and "black box" knowledge in yet others. We won't care. There is no inherent advantage to the aspects we call "emotion" unless it translates into a technical method for achieving these which proves more effective and efficient than the others.

But... and here's the "but"....

Nobody really knows what emotion is. Perhaps the things a robot would know, are analogs of emotion. Perhaps if a neural network learns enough, it displays states and decision-making which could be interpreted as emotion. We might never know if it freed that captive or didn't kill that child because of "emotion" or because of calculation, or because it weighed up the military and PR pros and cons.... and it might not matter.

Emotions give no magic. If we want certainty of the robots conduct, then that's the goal, and whatever works, works

Answer 10

Empathy

"It is not enough to conquer; one must know how to seduce" (Voltaire). Emotions would help a robot develop a framework to understand how to interact with humans, both allies and enemies.

Without empathy the machine has no reason to not kill off everything it faces. With empathy it might be capable of accepting the surrender of enemies. It will make it more apt to use non-lethal force against potentially non-combatant unknowns, or to take extra effort to avoid harming non-combatants.

Intuition

We as humans rely a great deal on intuition, gut instincts, etc. Emotions play a heavy role in that. In an effort to give your AIs some sort of intuition, they may decide emotions are a necessary component to help bridge the gap between what you know and what you don't -- the unknowns that are intuition.

Right vs. Wrong

While you can program a system to know all the rules of engagement, all of the laws, all of the hard-and-fast boundaries, that's not the same as knowing right from wrong.

Give a robot emotions, though, and you give it a tool to help differentiate right from wrong in those ambiguous situations that aren't clearly defined by the rules.

Fear

Fear is an emotion. It is also a valuable survival tool. Rather than program a robot in all the ways it can be harmed beyond repair, give it fear. Fear means it'll try to find a safer way. It may take different risks that reduce costly repairs or reduce the risk that the robot is destroyed before mission completion.

Answer 11

Emotions are how you program an intelligent machine.

It is the same reasons mother nature gave rats and goldfish emotions it is part of how we learn, and helps prioritize actions, pain in particular will very useful, if you keep doing something that damages you, a signal telling you that it is damaging you so you should not do that is very useful, that is all pain is. The more expensive the robot the more being able to feel pain will be an advantage, that way it is far less likely to damage itself for no reason.

Even subtle emotions like discomfort can be useful, if the robot feels discomfort every time civilians are harmed it will be far better at avoiding civilian casualties than any pre-programming could do. Likewise if it gets satisfaction from dead enemies it will get better at killing them.

the simplest emotions are just learning tags, you take an action and its results and tag it with one of two thing, keep doing that (pleasure) or stop doing that (discomfort). That's how to program dynamic learning. Emotions are basically the only way to program such a complex decision making machine as a combat robot, one that cannot possibly be pre-programmed but has to learn and respond to intelligent opponents. Otherwise your robot has the intelligence of the lowest insect at best.

Answer 12

Adding personnality and emotions to a powerful weapon seems a very risky thing to do. You make it unreliable and less efficient. The only (vague) advantage would be some unpredictability that would create some surprize effects, but I strongly doubt this would be efficient.

Emotions are not an advantage for beating us at chess or poker, while would it be in a war context ? Even if the reality of war has magnitude more complexity that such games, I cannot imagine any emerging property of such environments where non-rationale behaviour would be helpful.