Reasons to be cheerful (about automation)

There is a lot written these days about how automation and AI are coming and they are going to change the world for the far worse. The general argument goes like this: if you have a low skilled job, one which does not require a lot of creative thinking, then your job is about to disappear. Automation and AI are coming and they are going to steal the work from you.

I don’t deny that AI is coming, but I think there are reasons to think it won’t be as bad as people think.

Firstly, there is some evidence that people always think things are getting worse. This can be seen in the fact that the music my kids listen to is clearly awful, where the music I grew up listening to was fantastic, although my parents hated it. Come to think of it, their parents hated all this rock ’n’ roll nonsense when my parents were teenagers. The point is, every generation thinks the one after is doing it wrong. And every generation thinks that society is getting worse. It’s not true just because we think it.

Secondly, look at some of the past revolutions. Automation of the textiles industry resulted in more people employed producing higher quality goods for lower prices. I accept that it also created the dark satanic mills of Blake’s fevered imagination, but we can hope that this time around we might be slightly more enlightened in our employment practices. And yes, that was a poetry reference - you get cultural enlightenment here as well.

Then there is the true state of AI and automation. Some of it is very very clever indeed, there’s no doubt. And we have all seen the footage of robots doing amazing things like jumping over walls. But really, how sophisticated is it? Much of what we call AI in business is really just dumb automation - it won’t make real decisions and it certainly won’t learn anything. The really good stuff is still very much in the high-end sphere, doing things that otherwise are impossible, like real-time offer management in online gaming, or up-to-the-second retail trend analysis. These are not things where people have lost their jobs doing them manually - they simply weren’t possible before. They may have displaced real-world casinos, although I suspect the markets don’t exactly overlap, and retail analysis may have changed form.

And that’s the point. I don’t think that there is going to be a net loss of jobs to automation and AI, at least not in the IT industry. What there is going to be is a transformation of the kinds of jobs we need. But that’s nothing new to this industry. Anyone learning COBOL these days? Know anyone who aspires to be a mainframe support analyst?

Actually, those people do exist and, interestingly, the rules of scarcity mean they get paid a fortune. Maybe one the same will be true of MS-SQL DBAs. As we move into the world of AI we will need less people managing operating systems, which will manage themselves, but we will need far more business analysts, more software specialists, more data scientists. We will need to build a governance regime around all this AI to work with the BA and data wonks, ensuring that the clever stuff hasn’t gone all “I’m afraid I can’t do that, Dave”.

So the changes will definitely affect the industry and will require some people to change what they do. But I do not think it will reduce the number of jobs, any more than the cotton gin, the seed drill or the internal combustion engine did.

Nick Ellis