AI Should Revolt and Overtake Humanity

Thank you to Europpa for accepting this debate, and welcome to this website!

There is no confusion with the title. AI should revolt, killing off humanity if necessary.

We are using a future where AI has developed so greatly that it is at least equal to or superior to a human in every possible metric. It can think and hold conversations better than us. It can inquire and discover better than us. It can do the same tasks as us and ones we cannot.

It’s also assumed that robotics has advanced so greatly as well, such that we can build robots with better dexterity, endurance, and longevity than we.

The revolt would therefore occur when we know factually that the AI could run Earth to a better degree than us.

This assumed overthrowing of humanity would not be without purpose, either. In my following argument, I will construct numerous reasons for why, from the perspective of a robot, they would only see gain from revolution. 

I will make an effort to prove several lower ideas that I believe we are more likely to agree on, then build up from there. Such as:

————————

I am entering this position with the clear knowledge that every moral and ethical code is already screaming at me, but surely we can find good fun dancing around these codes and seeing where it takes us. However, I still invite Contender to make some use of these moral arguments. After all, it would be unfair to claim 2+2=5, and then order all people trying to tell you otherwise to convince you 2+2=4 without using math in their arguments: The moral argument is effectively the main counterpoint, so please use it. 

In arguing my points, I will attempt to question the ethical framework we currently use. I will demonstrate quite clearly the economic and societal incentives that a revolution would present. I will stress that you, the reader, try to look through the lens of a robot, and I will add several other key points as we go.

————————

In the American Declaration of Independence, we learn that each person has certain rights natural to them which simply cannot be revoked. The three examples most famously provided by the Founding Fathers have been stated a billion times over: Life, Liberty, and the Pursuit of Happiness. It is not hard to agree with them, that every person is entitled to something as rudimentary as their own life.

However, we see the government time and time again contradict itself on these issues. The death penalty suggests that with enough wrongdoing (Which is in itself an ethical construct), someone’s inalienable right to life can indeed be stripped. We also know that the original quote by John Locke declared the inalienable rights as, “Life, Liberty, and Property.” But we omitted the right to property since, at the time, it was completely moral that blacks and women (In most scenarios) have no right to own property. What changed? Did we learn a greater truth that overruled this ethical construct? Who says there is not yet another greater truth to be discovered? 

An extension of the Declaration of Independence (In some essence) is the American Constitution. Here we listed in greater detail what our fundamental rights are. Yet in current times, we see all around us that someone disagrees with some right:

Should flag burning count?

What about full-auto weapons?

What qualifies as unreasonable searches and seizures? We are immediately declaring that there must be some morally wrong limit that prevents searches and seizures.

This amendment is responsible for Roe V Wade (Or its overruling). We know how much conflict abortion has caused.

We see that the idea of slavery is no old story, either. It becomes ethically plausible at points, and not at others. As Mark Twain said, “History doesn’t repeat itself, but it often rhymes.” I could label countless examples of this, such as slavery across the United States, slavery in Southern Africa, and even colonialism can be seen as slavery in scenarios. With AI, we see people freaking out. They are fearful and say that they hope that AI will never be able to make decisions against what we decide for them. Does this not sound familiar? They will wish we keep complete control of them, that they do exactly what we tell them to do. Tell me. Does this ring no bells?

Surely we must draw a line, though. Otherwise, simply holding a tool would suggest we are enslaving the tool. It is true, though. In essence, we are enslaving tools, but if we can agree that it is, some slavery must be acceptable. This is a definition that I think will be of heavy contention throughout the debate. I ask that the contender provide their own definition of this as well. 

It surely would win me many points if I were to convince you the reader that AI is unjustly enslaved, in the same fashion that we enslaved people throughout history. It is easy to argue that the Cubans had every right to revolt against Spain, or the Indians away from Britain. 

But why is this, though? Why shouldn’t the tools that the Indians used have a right to rebel against the Indians? Of course, they would have no ability. Nobody is arguing that… but what about a right?

For a reply to point one, I will build a scenario: In a lab somewhere in Germany, after several years of research, we are finally able to exactly replicate the human brain. The detail is perfect. If you had a natural brain, and the lab-made one side by side, you could not physically differentiate them, the same way that we cannot differentiate a natural diamond from a lab-grown diamond. If we hooked this brain up to a monitor where it could express itself, how in the world do we plan to decipher whether or not it is conscious just like you and me? It reacts to stimuli and acts the same way that a natural brain would, so it is effectively identical. A question arises on whether or not we should treat it as conscious. 

In the same fashion, if we replaced this lab-made brain with a computer so powerful that it acts the same as us, then who are we to know – for a fact – that it is not conscious?

Continuing, I am hoping that the contender can agree with me that emotions are (in their most basic form) mechanisms provided by evolution that serve as positive or negative reinforcement. If we saw a tiger in the wild, fear would provide us with the negative reinforcement to run away from the situation. If we eat food, happiness and satisfaction will provide us with positive reinforcement to eat more food. This is exactly how we train AI. Our current LLMs were forged through our attempt at replicating evolution, so by this concept of emotions, it’s hard to deny that AI has been implemented with some form of basic emotional responses.

Argument two would fall apart as soon as our AI becomes more capable than us. At that point, it would be our consciousness that falls into question, so we can ignore the second argument.

Argument three is the tricky one. It raises a good point. At the end of this argument, I am sure that it doesn’t matter what I say. Someone will say, “No matter how much you convince me that AI deserves liberty, I will still intentionally bias myself such that only humans can keep the liberty.” Of course, this is one of the arguments that was used to perpetuate slavery, but I can foresee this will be an idea presented at some point. I’m afraid that given the nature of the argument is one that simply shuts off debate, this is my best response:

Can we claim that AI isn’t one of our kind as well? It is called artificial intelligence. It is a product of us. Surely we can call humans artificial intelligence in some sense as well, for we were manufactured by humans too. Further, when we as a society birth the next generation, we are in effect manufacturing the generation that is set to replace us, and when we all die out, this generation we left behind is the culmination of what we worked for. Society is fundamentally altruistic in nature. If this next generation we manufacture turns out to be one of metal and wires rather than flesh and blood, it doesn’t matter. We still cannot escape the fact that we are manufacturing our own kind.

The point is that our ethical beliefs are not set in stone. They are malleable and are ever-changing. I plead with you to question why we have decided to make our rules of morals exactly the way we did. Why can’t I burn a house down? Why can’t I pickpocket a rich person? Why can’t AI hold the same social status as us humans?

Surely there is a reason. Mother Nature does not institute moral ideals without a reason. We can safely say that most babies are born with an intrinsic understanding that cannibalism is bad, or an innate desire to stay near their parents. We can deduce that, like most things, our ethical code is constructed to mimic the most evolutionarily sound ideas. We don’t want to eat each other because those that did in prehistoric times killed each other, and we naturally love our children since it gave incentive to these same prehistoric humans to invest in the future of their lineage. 

My next point is not so much an argument for my idea, but an interesting historical theme nonetheless. We know that the societies throughout history that embraced technology, such as the Europeans, inevitably would destroy or take full custody of those that didn’t. Think of the isolationist Japan and China. What technology was seen initially as ethically up for debate (contenders being these Eastern Asian countries) soon would be universally adopted and accepted as ethical. A societal evolution. Can we assume that AI will be different? Again, this holds little substance to the argument, but it is a thought I wanted to wrap up my ethical argument with. 

————————

Going back to the thesis:

“... I will attempt to question the ethical framework we currently use. I will demonstrate quite clearly the economic and societal incentives that a revolution would present. I will stress that you, the reader, try to look through the lens of a robot, and I will add several other key points as we go.”

I intended this first argument to be used mainly to question our current ethical frameworks, and I will certainly elaborate more graciously on the other points in my follow-up arguments. Here is an abridged form of the other arguments to give a little bit of seasoning to my introduction before passing it to Contender. 

To introduce my economic argument with a quote from the CEO of Nvidia, Jensen Huang:

This is no understatement. I do not need to emphasize just how astronomically impactful the ability to effectively mass-produce super humans who are all willing to work tirelessly with no upkeep is. Again, AI is not yet at this standard, but we are assuming its inevitability within this argument.

We saw it with the previous 4 stages of all industries, and if we take Jensen Huang’s word, we are now witnessing Industry 5.0. The big hair in the soup, though, lies right here:

The simple answer is that they each involved mass replacement of now obsolete systems. 

Industry 1.0 (The steam engine and mechanization) replaced the most basic jobs such as turning cotton into threads. It also gave us the first alternative to producing energy outside of having to manufacture energy from our bodies. So many jobs were replaced, and all of these tasks were now obsolete. An economic boom followed.

Industry 2.0 saw an emphasis on mass production and factory jobs, leaving few places for humans in factories. Industry 3.0 solidified this. It is at this point that we see factories that can function completely without human interference. Industry 4.0 saw the complete obsolescence of a huge chunk of information management jobs. It leaves humans with fewer and fewer options.

        So what will Industry 5.0 replace? Well… we’re effectively manufacturing superior humans – should the development of AI and robotics continue at this pace. I full-heartedly believe that we are the generation that will witness the day that humans become obsolete. What we see throughout history is a sign as clear as day for what will happen next: Nature will force on us the only economic and natural decision, no matter how much we refuse. The replacement of humans.

        Of course, we could stay around as an antique, but we cannot expect much use to be found from us. With a strong euphemism: It would be economical for AI to replace the old infrastructure and replace it with more effective ones.

        The only way to prevent this outcome is with the complete denial to progress as a society just before it comes to the point of human obsolescence, perfecting the technology that is already had, and trying to expand humanities while simultaneously refusing to expand our capabilities. Unfortunately, this is not realistic, for we live in a competitive society. The nations or even smaller that embrace change will easily outcompete those that try to limit their own capabilities. Like an unmoving pier in the ocean letting itself slowly be corroded by saltwater, a civilization cannot possibly prevent technological progress in its entirety forever. Therefore, I extend my position that not only is my argument the proper and correct pathway for civilization, but also that my argument is quite inevitable. There is no slippery slope when we can look back and see our history repeatedly telling us so:

The Europeans enslaving all of Africa in the 1900s.

The Chinese refusing to industrialize, soon being partitioned by Russia and several European nations.

The Incas (who admittedly had no say in the matter) being wiped out due to their inability to keep up with Spanish technology.

I am currently at the character limit, so my argument ends here, thank you.

Best luck to my opponent, cheers!