Why AI And Robots Are Triggering The Next Industrial Revolution
Every month, we will take a look at one of the big emerging trends in the business (and, for our purposes, investing) world. First, we will examine the trend, and then spend the month looking at stocks that will benefit - or be challenged by - that trend going forward. Hopefully, we can identify some attractive stocks to consider, as well as some to stay away from.
For December, we are going to focus on one of the most covered, hyped - and probably most important - trends invading the business world today. And that is the emergence of robotics and artificial intelligence (AI) across nearly all business sectors, from industrial to service to consumer products.
So what is this important new trend and how will it effect business - and, by proxy, investment - going forward?
Building Upon Prior Mega-Trends
Robotics and AI are simply the next phase of a trend that has been developing since the 1700's - that of machines becoming sophisticated enough to replace human labor to complete work tasks faster, cheaper, more safely, and with better quality.
In essence, throughout history machines have increased productivity, allowing us to accomplish the same amount of work with less human effort.
Before the first industrial revolution, humans did most work manually. Sure, there were tools, and some primitive powered machines like water and windmills. But the key point is that these tools and machines were collaborative. That is, they were designed to assist humans to make work tasks easier or faster - not to replace human labor entirely. Productivity gains were modest but meaningful.
In the late 1700's, this began to change. Britain's trading empire of the 17th and 18th centuries created substantial wealth, freeing up time and labor to develop new technologies to increase productivity. Inventors of the day harnessed water and steam power to create powered machines that could do the work of 10, 20, 50 or more people in the same amount of time with the same or better quality. The textile industry - one of the largest of the time - saw productivity per worker increase by over 50 times as inventions such as the cotton gin (1793), the power loom (1785), and cotton spinners did previously manual work in a fraction of the time. The development of the steam engine was a huge invention that led to numerous advances. One of these was the rise of machine tools like lathes, planers, and boring machines - all dramatically increasing productivity of smiths, carpenters, and other craftsmen.
This was known as the first industrial revolution.
In the second industrial revolution, electrification and the internal combustion engine led to the assembly line, and advances allowing cheaper mass production of steel led to the build-out of railroads, allowing goods to be distributed at bulk across long distances. Instead of artisans or craftsman completing products by hand, production across nearly all industrial segments was converted to factories in which relatively unskilled labor could learn one task and create products far more cheaply at scale.
The third industrial revolution is one often referred to as the "digital age". It started with the proliferation of programmable computers in the 1970's, and kicked into overdrive with the build-out of the Internet in the 1990's. Now, software could be used to automate all kinds of formerly human tasks. ATMs used software and connectivity to essentially replace the tasks of a human bank teller. Phone and later Internet-based stock brokers replaced the work of human brokers, automatically matching and executing trades far more efficiently. Automated checkout machines could do the work of a cashier. And so on...
Robotics and AI represents what many are calling the fourth industrial revolution, although some see it as an extension of the third. How does this differ from what has come before?
The Fourth Industrial Revolution
There are two key points that distinguish robotics and AI from the simple automation that characterized the first 3 industrial revolutions.
For one, robots (both hardware and software-based) can react to a wide array of external stimuli to alter internal decisions on how they operate. For example, a consumer robotic vacuum has sensors that allow it to work around obstacles and "remember" where it has already vacuumed. This differs from the automation of the past, where machines were designed to do a very limited number of tasks in one specific way, with simple error detection and handling being the only decision making.
The second big difference is that AI is designed in a general way and then "taught" based on "training" from millions of sample situations. This differs from automation of the past which was purpose-designed by humans. The advantage here is that AI-based bots can learn new things without human intervention. For example, consider a software customer service bot. While initially, it may only be to automate basic queries, over time it can be "trained" on the proper responses to more complex queries, expanding its usefulness.
The implications of these two factors are pretty immense. Over these industrial revolutions, we've seen machines be able to do the work first of manual laborers (early powered machines), then of process-based craftsmen (assembly line), then of simple decision-based mental work (software). With robotics and AI, machines will be able to automate tasks that require traits once considered uniquely human, such as learning on the fly, or making complex decisions based on hundreds of external stimuli.
The Implications of Robotics and AI
What are some of the human tasks that will be replicated through the rise of these trends?
Here are just a few examples:
- Driving. Already numerous companies are in the latter stages of developing automated vehicles that can navigate with no human intervention whatsoever. This opens up the possibility of automating such currently human jobs as truck drivers and taxi drivers.
- Customer Service Positions: We already saw the start of this during the third age, but it will only accelerate. AI-driven software bots will be able to handle more and more customer support queries. Automated checkout solutions, including zero-friction sensor-based ones, will continue to replace store clerks.
- Public Safety: Policemen, firefighters, soldiers, and rescue workers have extremely dangerous jobs. This makes the area a natural target for robotics. Robots can already access areas inaccessible to humans, do aerial inspection and reconnaissance, disable minefields, conduct bombing raids, etc. More robots capable of even more tasks in these areas are inevitable.
- Research Assistants: With the near-complete electronification of documents, and AI developed enough to understand context, new research bots are emerging that can scan millions of documents and produce briefs in a fraction of the time and cost of an army of research assistants. The ramifications, particularly in research-heavy fields like legal, are immense.
- Pharmacist: Advances in the mechanical abilities of robotics, sensors, and the AI to understand the data has allowed some hospitals to implement fully robotic pharmacies that accept prescriptions electronically and fill them rapidly and accurately.
- Industrial Production: Long an area where automation has proliferated, the advancements of sensor and AI technology have even led to robots taking formerly human positions, such as inspection and quality analysis. Warehouses are becoming increasingly automated with fetching, packing, and shipping robots.
- Surgery: Robotic systems already assist surgeons by allowing more precise operations and enhanced visualization, reducing the invasiveness of surgery and improving patient outcomes. In the future, these systems may be able to work more and more autonomously.
- Housework: Probably my favorite area, several consumer robotics firms have developed devices that can automatically handle many of the dreary home maintenance tasks like vacuuming, sweeping, mopping... even gutter cleaning and lawn mowing! How long will it be before we have automated pressure washing and painting bots? Probably not long.
These are just some of the advances. As you can imagine, with sufficiently advanced mechanical abilities and trained AI, there aren't a whole lot of human tasks that robots won't be able to conceivably tackle.
The Investment Angle
Look at some of these projection statistics:
- Worldwide spending on robotics and related systems is estimated to grow from $71 billion in 2015 to over $135 billion by 2019, a compound annual growth rate of 17%.
- Industrial robot sales are projected to grow from about 250,000 units in 2015 to over 410,000 units in 2019, a compound annual growth rate exceeding 13%.
- Medical robotic system sales are projected to growth from $6.4 billion in 2016 to over $20 billion by 2022, a compound growth rate of nearly 22%.
- Consumer robots are still early in the adoption phase but have big potential. The market for just robotic vacuum and mopping systems alone is expected to grow from 4 million units in 2015 to 30 million by 2019, a robust 65% annual compound growth rate.
And these are just a few potential markets.
Remember, robots can work 24/7, produce perfectly consistent, fast, and accurate work, do not require wages or benefits and do not get sick or bored. It is in the interest of almost any management team to investigate the application of robotics to solve business problems. The social implications of these changes will not be unlike the social implications of the first 3 industrial revolutions, and we are seeing them already.
But this site isn't about the social implications. The investment implication is that a lot of money is going to go to the companies that produce robotic solutions to the needs of various industries.
Conclusion
AI-driven robotics are an irresistible force that is just now beginning to gain traction. It is a clear continuation of the industrial revolution trends that started over 200 years ago. Like the cotton gin, the assembly line, or programmable computers, it will lead to the replacement of human labor by machine labor in ever-expanding industries.
The investment implications are immense. A lot of money will go to the companies that can provide robotic solutions to produce goods and services cheaper, better, and faster.
In December, we will take a look at several of these companies that are emerging as major providers of these robotic solutions (and/or the pieces needed to create them). Stay tuned to MagicDiligence ...
more