Why your job is threatened by Artificial Intelligence

AI will eliminate jobs. Will new ones be created in time?
AI will eliminate jobs. Will new ones be created in time?

For a long time, AI was the stuff of Hollywood movies, usually cast in a villain’s role. Usually, the machine was endowed with megalomaniacal tendencies, intent on wiping out humanity. While this made for good entertainment, the reality of AI was a completely different story.

But the highly visible, recent successes of AI and machine learning (for example Google’s self-driving car) have led many commentators to re-evaluate the possible impact of Artificial Intelligence on society and people’s jobs.

In 2013, Carl Frey and Michael Osborne from Oxford University published a paper that reviewed 702 different jobs and estimated that 47 percent of them would be threatened by computerisation. This led to a flurry of articles about this possible threat, some saying that the number is too high, others too low, especially when confronted by recent advances in Artificial Intelligence.

J.P. Gownder at Forrester published a study more recently that estimated 25% of jobs would change and that increasing use of AI would cause a net reduction in jobs of 7%. Wired is taking a more balanced view that as with all technological advances, some jobs would be lost and others would be created. Even the World Economic Forum has pitched in, calling for world governments to address the new requirements with appropriate skills for people.

Some others have taken the view that there are some things that AI will never be able to do. Hampus Jakobsson, writing for The Next Web, optimistically identifies five uniquely human-centric jobs that machines will never be able to so, including Salesperson and Product Psychologist. Similarly, Nigel Webb maintains that AI will not be able to create more engaging advertising. He makes the appeal that engagement in advertising is primarily an emotional response.

Even where AI is being deployed today, there seems to be a form of double-think going on. For example, Clara is a software driven personal assistant, that can schedule appointments simply by being included on the email distribution. In a recent interview, Maran Nelson, the CEO of Clara Labs who created Clara, stated that “the intention is not to replace humans with tech, just remove more tedious human jobs”.

This particular example goes to the heart of the matter. Clara takes one time-consuming tedious task (but tricky to accomplish) and automated it. This is exactly what computerization and in fact technology has been doing for the last several decades, if not longer. So while their intention is not to replace people, that will in fact be the ultimate outcome.

“Everything that can be automated, will.” – Shoshana Zuboff, Professor of Business Administration, Harvard Business School (Retired)

Prof. Zuboff wrote this in the 1980s, in relation to the advancing computerization. When we follow this logically it means that any task is subject to automation, to a varying degree of difficulty and feasibility. When we look at our jobs, we find that each one is made up of a combination of tasks and responsibilities. Some of these tasks are complex, requiring lots of skill and intelligence to complete. Others are boring and repetitive. Commercial AI will attack these latter tasks first and as researchers learn more they will increasingly attack the former.

Practically every job in existence today has been affected by technological progress in some fashion. This shows no signs of stopping. Every job will be affected in some way by AI, whether it’s the elimination of parts of the job or the wholesale replacement of the job by a range of task-oriented AIs. It’s important to note that this does not require legions of general-AI robots or machines. It only requires legions of narrow-AI, some of which are already with us today.

To claim that some jobs are immune to AI is to ignore the history of technological progress. This progress is not going to stop, so we might as well accept it and examine how the nature of work will change in the next few decades and it’s effect on people.

In an interview back in March, Hilary Schaub (a Personal Assistant to a Vice President) used another automated scheduling assistant called Amy (from xdotai). She said that Amy eliminated a boring part of her job, but felt that her job was safe. What she fails to recognize is that there will be a Bob, Chris and Daisy coming along that will do other parts of her job. Pretty soon, her job will be redundant, dead from the cuts from a thousand AIs.

Artificial Intelligence is going to have a profound impact on jobs and public policy makers should be aware that “creating jobs” won’t be as important in the future as “creating employment opportunities”. The WEF call to action is sound and timely. Unfortunately it will probably be ignored until the problem is acute.

The clock has been running for a while and it has just started ticking faster.


Frey & Osborne (Paper/PDF): The Future Of Employment: How Susceptible Are Jobs To Computerisation?

JP Gownder (Forrester Blog): Robots Won’t Steal All The Jobs — But They’ll Transform The Way We Work

Cade Metz (Wired): Robots Will Steal Our Jobs, But They’ll Give Us New Ones

Sudan & Yadunath (WEF): Are we heading towards a jobless future?

Jakobsson (The Next Web): 5 tech jobs machines will never be able to do

Webb (Campaign): Why artificial intelligence will not create more engaging ads

Nelson (USA Today): Clara is applying to be your virtual personal assistant, no benefits required

Professor Shoshana Zuboff (Personal Site)

Schaub & West: Should I worry about Amy the AI robot taking my job?


Using Deep Learning to Analyze Genetic Mutations

Tiny part of a strand of DNA

Here’s a great application of machine learning: looking for genetic mutations that could cause disease.

The Human Genome Project mapped the sequences of base pairs in DNA during the 1990s. However, it generated an enormous amount of data, and now teams are applying machine learning techniques in order to analyze the genome. The hope is to be able to identify those mutations in the genome sequence that cause diseases and that could lead to early intervention and perhaps new treatments in the longer term.

It’s nice to see big data and machine learning applications in other areas than trying to improve advertising!

The interview below is with Brendan Frey, CEO of Deep Genomics in Canada and gives a nice overview of what that team is doing and how they are hoping to leverage machine learning.


Using deep learning to analyze genetic mutations: an interview with Brendan Frey.

Deep Genomics Website

DNA Animation: brian0918™ (Own work) [Public domain], via Wikimedia Commons

Deep Blue’s Child: Computer Learns To Play Chess

Remember Deep Blue? Back in 1996, IBM created a chess-playing computer that beat the then world Champion Gary Kasparov. That machine’s impressive results were the result of it being able to evaluate over 200 million positions per second. Naturally, Kasparov didn’t take that approach. So it was really a triumph of brawn over brains. Since then chess programs have gotten better, though usually because the programmers encoded strategies in the software.

Fast forward to today and playing chess is still a worthy problem to tackle. This time however the emphasis is on learning rather than brute force evaluation. Matthew Lai has created a system that learns the rules and discovers patterns from analysing previous games by humans and machines. Much like a human chess player does. It also played itself, trying to improve it’s techniques.

Using an Artificial Neural Network learning system has been a successful approach and Lai’s system has now reached International Master Level. It’s quite possible that this AI could even discover new strategies and tactics for winning as it continues to learn.


Artificial Intelligence Is Taking Computer Chess Beyond Brute Force | Popular Science.

Deep Blue Chess Computer (Wikipedia)

Toyota Pursuing Alternative AI Car Strategy

Toyota is investing $50m into driver-assisted Artificial Intelligence, in partnership with MIT and Stanford, according to Fortune. The research will be focused on helping human drivers avoid accidents by adding AI to their cars.

This is in contrast to the seemingly much more ambitious self-driving cars being pursued by Google and others. Although Toyota itself demonstrated it’s own self-driving Lexus in 2013, it seems that they also believe that people will still want to drive their own cars. By investing in AI technology to augment human drivers, Toyota seems to be hedging its bets.

There is still a long way to go before we will see fully autonomous cars and trucks commonly on the roads. The fruits of this research may be installed in your car much sooner than that.


Toyota partners with MIT and Stanford on artificial intelligence – Fortune

Toyota sneak previews self-drive car ahead of tech show