I love automation. I love it for the efficiency it brings. I love it for the freedom it brings to humanity. I’ve observed over the years, that not everyone shares my passion for this subject, so I thought I’d write a brief defense of automation and offer a few cautions.

I tried to think back to my first real encounter with the power of automation. It goes back to my childhood. Once upon a time, I was terrible at math. Well, that’s not completely true. I was slow at math. I usually knew how to solve problems, but it took me too long and required my full attention. I usually ran out of time on tests, and therefore got bad grades.

Upon entering junior high school, I was given the freedom to use a calculator for math. Before that, using a calculator was considered cheating. In algebra and beyond, I could use my calculator for simple calculations and use my mind for abstract problem-solving. For the first time, I felt smart. Math became easy.

To this day, I find mental math challenging. It takes my full attention to calculate a tip. I appreciate it when suggested tip percentages are presented at a point of sale. It just makes my life easier.

I’d rather interact with a machine than a human that acts like a machine.

Some bemoan the automation of customer-facing roles such as a bank teller or a check-out clerk. For me, those interactions are always very transactional. They aren’t relational at all. In fact, due to my introverted personality, I really don’t enjoy transactional human interactions. I’d prefer to avoid them altogether and just interact with the machines.

Of course, if the machines go haywire, or if I have a need that the machines can’t handle, I do appreciate the support of actual humans that can solve problems.

I’d rather interact with a human than a machine that tries to act like a human.

I don’t talk to Siri, Cortana, Alexa, or any of their friends. I just can’t do it. It feels completely unnatural to me. For me, speech is for conversations with humans. Computers can’t carry on a conversation. They can execute transactions and follow commands. I already avoid transactional conversations with humans. Even though I am a leader, I don’t go around issuing verbal commands to my employees. We have conversations.

I realize that for some, a speech interface provides the accessibility that they wouldn’t otherwise have. For those individuals, I’m happy for them. It’s just not for me.

The ideal

Machines should do machine things. Humans should do human things. When we get that wrong, it’s a problem.

Machines are great at performing repetitive tasks with speed and precision. Humans are awful at that.

Humans are great at being creative, strategic, caring, relational, and funny. Humans are complex problem solvers, leaders, and visionaries. Machines are awful at all of that.

How to automate incorrectly:

There is a lot of automation going on in our world, and not all of it is good. We mess up in two ways:

1. We automated the wrong thing. We should automate transactional interactions, not relational ones. If we remove the relational element, then we remove the human element, and therefore dehumanize our organizations. Let’s not do that.
2. We automate the right thing, but poorly. Machines are fallible because they are made by fallible humans. The first generation of automation is usually clunky and error-prone. That’s okay, but just don’t leave it there. Keep improving it. And please keep supporting it with actual humans while it’s under development.

I’m going to end this article with a reference to the 1983 film, WarGames. As a child of the ’80s and a computer geek, this film has a special place in my heart. WarGames starts with an innocent computer game with an artificially intelligent computer, then quickly escalates to a global thermonuclear war.

At the climax of the film, the computer’s developer, Professor Falken, asks an important question: “But does it make any sense?”

We always need to be asking that question. There’s a lot of talk about data-driven decision-making. That’s good, but it might be better to have data-informed decision-making, especially when the stakes are high.

After plenty of disagreement and anxiety, Professor Falken saves the day with his final plea: “General, you are listening to a machine! Do the world a favor and don’t act like one.”

Check out the full scene here:

Podcast: Play in new window | Download