Yup, I said it: Robots and artificial intelligence are not going to take over the world. Am I a pariah now? What’s more, I think AI alarmists have it all wrong. How can that be?! Just hear me out. Here’s why AI won’t take over the world:
AI Is Not Like Previous Technologies
First, before the haters start to hate, let me get this out of the way in the beginning. Yes, AI will likely be vastly smarter than humans, and also robots will be vastly more capable than humans, physically and mentally. What’s more, it will be a challenge to control general AI and to ensure it has the proper values. Oh, and all this may all very well happen within my lifetime, if not our children’s lifetime. The singularity will be here before we know it.
I am not debating that at all. AI is going to be smarter than you can imagine, and it will be here faster than you realize! But that doesn’t mean it can take over the world, or even take all of our jobs. Here’s what I’m talking about:
First, A Legal Argument
AI and robots have no legal rights. They can’t own property, they can’t open a bank account, they can’t start a company, or own one, and they can’t be represented in the courts to sue for their rights either. Because they have no rights – any more than your sock does.
They can’t vote, pass laws, or become President. They can’t even sign a contract or take a loan or own or rent property. These may seem like small parts of your life, but these are the types of legal rights that allow humans to accumulate and hold power in a civil society. These rights are behind everything in our world.
AI would have to militarize and take control over the physical landscape in order to upend our legal system (like in Terminator). That is of course what AI alarmists propose could happen – a military takeover. So let’s get back to that later. But for now, let’s pretend there will be no war between humans and the machines.
Short of war, the only way to control the world is through these types of legal rights. And only humans have them. AI may become smarter than we can imagine, but ultimately it can only use those brains in an advisory capacity to the fools that control things – the humans with legal rights.
I also don’t believe that AI could likely manipulate people into voluntarily giving up those rights. People have a strong desire for both power and self determination, and even if AI is vastly smarter, people are not getting dumber.
Second, An Economics Argument
Here’s where I think most AI alarmists really get it wrong. AI and their robot companions will be expensive technology. And that means they won’t be used everywhere. They will displace a lot of humans. Some areas may deteriorate economically (like Detroit). But AI and robots certainly won’t replace ALL humans – or even most of them.
Let me give an analogy. Tractors are a major technological advancement for farming. They’ve been around for decades. They can save both time and money. Yet, even today most farms on Earth don’t use them, and probably never will. Yes, most.
That’s because tractors are too expensive for a small farm, and believe it or not, the vast majority of farms in the world are small. Africa, Latin America, and Asia are full of small farms that will never use a tractor. They work with human and animal muscle – a clearly inferior technology – because it is cheaper. Maybe the folks in Silicon Valley don’t see too many small farms outside of their cubicles. But perhaps they need to get out more.
My family and I are currently living in Bolivia. Here is an example of what I’m talking about. This is a small farm that costs about $10,000.
A $100,000 John Deere tractor on this land would never make sense economically. Not even a used, beat-up, small $10,000 tractor would make sense on this land. Believe it or not, THIS is what most farm land in the world is like.
My point is that just because a superior technology exists doesn’t mean it will be widely adopted, if it is too expensive for real-world applications. Even tractors have not been widely adopted throughout the world. In fact, tractors only replaced humans in places where humans can earn more money doing other things – like the United States.
AI Is A Technology Of Wealth
There are all kinds of wonderful technologies that are not widely used outside of the wealthiest corners of the world (North America, Europe, Japan, etc).
AI and robots will be expensive, and will largely be exclusively used by the people and organizations that can afford the expensive technology – like large companies and wealthy consumers in the U.S. and Europe. If you think that’s a world takeover, you need to get out more. The world is much bigger than that.
Humans, On The Other Hand, Are Cheap
AI and robots will ALWAYS be more expensive than humans. Why is that? Because the marginal cost of a human is only what it takes to stay alive – the cost of shelter, food, and water. The cost of robots and AI on the other hand will always be higher. They require electricity, wiring, rare earth metals, and warehouses full of replacement parts. Humans, on the other hand, are self-repairing. You may not be happy about taking a pay cut, but you can do it and still stay alive, and even enjoy life. A robot is not so flexible.
We’re All In It Together
Why is this important? Let’s think through what happens when AI and robots take all of our jobs. First, companies will hire robots and fire their humans. If it’s just a few companies, then no big deal, right? Some people lose their jobs, adjust, and find other work. But if it happens across all industries like people are predicting, there will be millions upon millions of people unemployed.
All those unemployed will be feeling the pinch, so they won’t go out and spend as much money. The economy will crash and demand for goods and services will too. That means prices will go down, and life will get cheaper.
But the labor market is not a vacuum. Businesses will be impacted too. How will those same companies that fired their humans then react when facing lower prices and reduced demand for their goods and services? Some companies will be forced to put their robot orders on hold. Others may go out of business entirely (laying off their robots). Others may even hire back their human workers, if the humans are willing to work for less. Since humans cost less, they can do this.
Robots will only replace humans if humans can continue earning money doing other things
Humans Win In A Race To The Bottom
Perhaps this becomes a downward spiral where humans are bidding against robots for jobs at lower wages. I don’t believe this will happen, but if it does, humans will win out rather quickly, as a less expensive technology. In fact, robots can only take our jobs if there is a robust economy that can afford to pay for the robots. A bad economy is bad for humans, but it’s even worse for robots.
Humans will always work, even if facing stagnant or decreasing wages. We have to work in order to eat, and we can do it for much less money than robots can. Robots on the other hand, don’t have to work.
If humans are all completely out of work, robots will be right in front of us in the unemployment line.
And what about a farmer in Bolivia? How many small farming jobs do you think will be lost to robots or AI in a place like rural Bolivia? Exactly zero! Thinking that robots will take “all of our jobs” sounds to me like a misunderstanding of economics and ignorance of the world beyond Silicon Valley.
Two Possible Paths
That leaves really only two possible paths I can imagine to our economic future with AI: First, that humans don’t widely adopt AI because it is too expensive for real-world applications, and humans can do things cheaper. This is what will happen in most of the world – Africa, the small towns and rural areas of Latin America, rural India, rural China, Indonesia, etc. This is where over three-quarters of the worlds’ population lives.
The other possible path is where humans and AI both find ways to work. An economic balance forms where enough humans are making enough money in some ways that the economy can support the expense of AI for many applications – but not all. Some humans take pay cuts. Lots of them have to learn new jobs or new skills or new ways of making money. But, they adapt. Other humans make a killing (money-wise). This is what I believe will happen in the U.S. and other wealthy countries. Without high human employment, robots and AI won’t be used much either.
I suppose there is that third possibility we still need to discuss – war. And that brings me to my third argument.
Finally, A Military Argument
So, you see that humans will always have some dramatic advantages over AI, being 1) that only humans have the legal rights to accumulate wealth and power and 2) that humans are much cheaper and will always have work because of it. So, what if AI decides to go rogue and take over the world militarily?
OK, how many armies have taken over the entire world, ever? That’s right – none. Has anyone ever gotten close? No, not really. Not even Hitler or Napoleon. Europe, maybe, but the WHOLE world is a BIG place. Nobody has even come close to taking over the entire world because it would require control and coordination of an ENORMOUS amounts of resources and land. The world is way too large and fragmented for one organization to control it all.
But, AI is going to be super smart! Way smarter than any humans! It will figure out a way!…. Yeah, I’ve heard it, just stick with me for a minute.
So what if some malevolent dictator or Silicon Valley executive decides to unleash an evil AI that will try to take over the world? Or the AI just decides to do it by itself. Yes, this will probably happen. What challenges could it possibly have against the foolish humans? Well, for one, it won’t be the only AI in the world.
Why AI Won’t Take Over The World
Terminator is a great movie, but it’s pure fantasy. It’s not like there will be one AI that decides to fight in one corner vs. all of naked humanity in the other corner.
There will be many, many generally intelligent AI, and they will all have different mission statements. Already there are hundreds of silicon valley companies that are working on AI. I’m sure the government is too. Plus the Russians and the Chinese. Yup, there is going to be a shipload of different AI.
Some of those AI may want to kill all humans, while others will want to protect the status quo. Some may work for the benefit of organizations like the U.S. government, and others will work for wealthy individuals. Some will even go rogue. They will all have different goals and different values. They also will have different strategies and different defenses.
Do I believe we could have wars involving AI, or even started by AI? Yes, of course, we will. We will always have wars. We’re human! And since AI will be a revolutionary technology, of course it will be involved in our wars! If you think we’re in a post-war world, then I have a bridge to sell you!
It Won’t Be AI vs. The World
Will one AI be able to secure all the resources required to take over the entire world before facing defenses from other AI? Of course not. That’s ludicrous. There will be strong resistance to prevent it. And defenses will probably begin before a malevolent AI can even take over all of France, much less the entire world. The world is too large and too fragmented to think otherwise.
Hitler’s army could perhaps have taken over the world during Napoleon’s day – with far superior technology and weapons. But Hitler’s army didn’t exist during Napoleon’s day. It existed in the 1940s, and it was met with equivalent resistance from several other armies with similar technology at that time.
Similarly, any AI that tries to take over the world in the future could certainly cause some damage and kill some people. But, it will be met with resistance from similar technologies of its time.
Could AI take over the entire world? Nonsense. Wars will happen, and the human story will go on.
If you’ve made it this far, thanks for the read! I don’t pretend to be an expert in AI technologies, but I do know a few things about economics, I’ve watched the History Channel, and I know how to identify bullshit when I see it! The scare stories about AI are mostly hype by people who are stuck in the weeds of the technology, amazing as it is. Change is coming for sure. AI will dramatically change our world, but it won’t destroy it.