Does Artificial Intelligence Steal Human Jobs?

The Four Job Categories

When it comes to artificial intelligence (AI), many anticipate a rosy future with various sources of new revenue, reduced expenses, and ultimately increased profits. Others worry about the jobs that might be lost to machines.

So, does AI steal human jobs? Or put another way, should we replace humans with AI?

Before answering these questions, we first need to categorize different types of jobs. I’ve devised a table below that divides them into four categories based on yes or no answers to two questions. The four cells describe who or what should perform a particular task that falls into that specific category. Jobs can be described as roles, and the tasks are the problems that need to be solved within those roles.

To be sure, the table is simplified for illustrative purposes and not mutually exclusive, collectively exhaustive (MECE). That said, it should give financial, technology, and management professionals plenty of food for thought.

1. Traditional Computer Programs and Other Technologies Mainly for Process Automation

This category includes but is not limited to certain trading, money wiring, settlement, clearing, and other operations at banks, trading venues, and investment management firms. In a strict sense, humans often have to be involved for technical, economic, and legal and regulatory reasons, among others. Some humans might resist streamlined processes without human intervention all the way down the line. They will be inclined to cling to jobs that can be done by machine.

2. AI, Traditional Computer Programs, and Other Technologies

Some jobs that may fall into this category include recommending web content or applications based on user preferences and past web or app behavior. AI results can leave room for interpretation. The consequences of decision making are not that critical or significant. Even traditional computer programs and other technologies can be applied. Results from such applications often provide more and better results than humans and at scale.

3. Humans

The jobs of corporate executives, politicians, or any other person who makes decisions based not only on objective facts and simple rules and principles but also on long-term perspectives and human values are among those in this category. Decision-making processes are usually one-off, non-automatic, and often have irreversible consequences. Human decisions are not necessarily based only on short-term, economic, and rational reasons. What look like knee-jerk or irrational responses at first glance may in fact be based on subtle calculations. Moreover, humans can have subjective opinions, applying varying time scales, and acting on complicated rules and principles that cannot be reduced to relatively simple algorithms. Unlike machines, humans can take responsibility for a result and understand the legal and ethical obligations.

4. AI and Humans

This is an area where humans and AI (machines) compete for the job. Humans can be replaced by machines if all the following conditions are met:

1. Machines offer a better solution than humans based on costs, output quantity and quality, and so forth.
2. There are no legal restrictions.
3. It is appropriate according to normal social conventions and there is no ethical obligation to do otherwise.

In other cases, humans and machines can work together. We can solve problems by referring to the (past) data and envisioning an often complex future state. Humans should be good at the latter: We are “teachers” who know and can define what is a correct or incorrect answer, or future state. We can also assume responsibility for decision making and its results. AI has mastered many things and solved various problems standardized by human beings, but in other ways it can be outthought by a toddler. It requires frequent human intervention.

Stock selection, portfolio management, client services, sales, and other jobs with human interaction can fall into this category. The artistic realm is another area where this human-machine collaboration has worked well, in the form of, say, AI-assisted computer graphics.

The Solution: Focus on What Only Humans Can Do and Do Well

To avoid losing our jobs to machines, we humans need to identify and focus on what only we humans can do and excel at. We need to remember that only humans can define each job, what it does or does not require, and whether it can be assigned to machines. Dividing jobs into sub-jobs and then categorizing these into these groups is something that only humans can do and should be good at.

Furthermore, humans can transform a job, redefining it and moving it from one category to another. This way, humans can and should maximize the value of machines so that we can focus on more meaningful, productive, and enjoyable activities. In the end, humans have feelings: These are often unstable and seemingly irrational. Machines, thankfully, do not have them and will do only the tasks that we humans can assign them.

Of course, AI — “machines” — are only as intelligent as the data it learns from, the models and techniques that are deployed, and the humans that are associated with it. Raw data itself, data cleaning, and knowledge and experience about how the data is generated, collected, processed, stored, and analyzed, do matter. Selecting an appropriate model is also important as is understanding the objective of the analysis. The role of even subjective expert human judgment based on knowledge and experience is critical as well.

For various legal, ethical, and economic reasons, not all human jobs should be replaced by machines. But humans equipped with machines, by using a combination of AI and human intelligence, will replace some jobs. AI may transform our businesses, but it is not the existential threat to human jobs that many of us fear. Rather, those human teams that successfully adapt to the evolving landscape will persevere. Those that don’t are likely to render themselves obsolete.

What it all boils down to is it is our job — we humans, not the machines — to study the board and make our move.

If you liked this post, don’t forget to subscribe to the Enterprising Investor.

All posts are the opinion of the author. As such, they should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute or the author’s employer.

Image credit: ©Getty Images/ AerialPerspective Works

Professional Learning for CFA Institute Members

CFA Institute members are empowered to self-determine and self-report professional learning (PL) credits earned, including content on Enterprising Investor. Members can record credits easily using their online PL tracker.

Yoshimasa Satoh, CFA

Yoshimasa Satoh, CFA, is a director at Nasdaq. He also sits on the board of CFA Society Japan and is a regular member of CFA Society Sydney. He has been in charge of multi-asset portfolio management, trading, technology, and data science research and development throughout his career. Previously, he served as a portfolio manager of quantitative investment strategies at Goldman Sachs Asset Management and other companies. He started his career at Nomura Research Institute, where he led Nomura Securities’ equity trading technology team. He earned the CFA Institute Certificate in ESG Investing and holds a bachelor’s and master’s degree of engineering from the University of Tsukuba.