The Robotic Revolution

We are nearing the point of seeing a robotic revolution. This is going to be like something we never saw before. The potential disruption is on a scale never seen before.

For this reason, we will drive into some of the basics to highlight what is going to happen.

Source

Multiple Learning Curves

What happens when you get a new job?

If you are like most, we have to learn from scratch. Even if we have some experience in the industry, the policies and procedures of this particular company are new. The learning curve can be enhanced if we are dealing with a geographic change where the entire landscape is different. At a minimum, we are going to have to learn about the other employees and customers.

When someone leaves a job, the process is repeated. This is why turnover is lethal to a company. Each person is coming in new (unless a rehire). When one walks out the door, all knowledge goes with that person.

This is not the case with robots. As opposed to having a new learning curve with each additional employee, robots learn once and are done with it.

By this I am not referring to an end to the learning process. Instead, I am focusing upon a particular task. Once the system learns it, there is no reason to cover it again.

This means that, on an individual task basis, there is only a single learning curve. If a robot is replaced, simply load the latest version of the software to the new bot and it is good to go.

Exponential Learning

Here is something I do not hear many average people talking about. When discussing the pace of automation, this is a crucial element.

Humans are single learning. That means if I learn something, that is only me. For someone else to learn it, I can share the knowledge but it is up to that individual to grasp it. This can be knowledge or the ability to do something.

If there are 100 tasks, we will find that each employee is adept at varying numbers. Some might have all 100 down while others are in the range of 20 or 30. Plus, as just mentioned, each are going to require learning personally.

Again, robots are different.

We see the exponential kick off in a simple way. Using the 100 task example, each robot can learn 1. That means we have 100 tasks in total. However, if this were humans, it might apply. For robots, we put it on steroids.

Each robot is learning a single task which is uploaded to the system. This means the database of knowledge is adept at 100 tasks. Therefore, when the 100 robots are updated, each are equipped with the ability to do all 100.

We suddenly have a total of 10,000 tasks.

What happens if we have 1,000 robots each learning a single task? Suddenly, we are to the point of 1 million.

As you can imagine, the numbers can get rather big.

Time Well Spent

This video made the rounds last week, including in one of my articles. It shows a robot making coffee.

It was something that garnered the attention of the AI world. To be fair, we have no idea if there was human operation involved or the bot was completely on its own. Nevertheless, it does exhibit how things can operate.

We were told it took 10 hours to train the bot to do this. It is something that might lead some to believe things will evolve very slowly. After all, it would then take 1,000 hours to learn 100 tasks.

That is true. The problem with this is we are still using linear thinking. The presumption is 1,000 hours for 100 tasks per robot. It might not be the case.

If we had 1,000 of these Figure robots, that would mean 1 hour per task. Suddenly, we have exponential time.

Let us frame the conversation in another manner:

How long does it take us to generate a human worker?

From conception there is 9 months to birth. After that, we see 18 years to adulthood. For many professions, we are dealing with an additional 4 years of school, putting us at around 23 years. Of course, we might be dealing with a genius or a total moron.

Nevertheless, after all that, we have 1 worker. The process is very similar for each employee an organization has. Certainly, there will be variance based upon the industry. A fast food worker might start at 16 years old; a lawyer at 24.

Just consider where these technologies will be if we only apply the gestation period of a human. Another 9 months could mean a radical uptick in the capabilities.

This is why we are nearing the robotic revolution. If the actuators and other mechanical components are working, the AI is going to be there. Sure, it might take a lot of input to start. However, none of that is wasted as the database of abilities grows. Then, it becomes exponential as more are built and put into operation.

It is going to be a radical shift in the labor market.

---

What Is Hive

Posted Using InLeo Alpha