AI is a good tool, but it won’t take over your job – yet

Have you heard this one before? That artificial intelligence, or robots, or some as-yet-undefined piece of technology is coming to take your job? It’s scary and anxiety provoking, and seems, with recent technological advancements, like it could be true. But is it?

The answer, as you might expect, is murky. The best guess from experts is that your job is probably safe – for now. But, depending on your career sector, your job will almost certainly see changes in the next few years, if it hasn’t already. Instead of thinking about AI as a great replacement of the human work force, think of AI as the harbingers of a great augmentation.

“A human and a computer working together as a team can actually be better than one of them working alone. It’s a good example of the whole being greater than the sum of their parts,” says Fred Popowich, professor of computing science at Simon Fraser University. Mr. Popowich says there are certain skills where AI excels, such as analyzing large reams of data quickly, or spotting patterns. But humans have the tech beat in other areas, like understanding tone or shifts in meaning. So, AI is often great at data analytics, but people are still needed for more cerebral work.

The advancements in AI are like any other technological shift, Mr. Popowich says, whether that’s advent of the printing press, the automobile, or the home computer. There are new skills humans will have to learn, and ways that our work will change, but it’s unlikely that AI will overtake office buildings across the country.

However, there are a few sectors that are particularly ripe for AI integration, says Mr. Popowich: finance, climate, and health. In fact, AI is already being used in these sectors, such as hospitals across Canada. The goal with this AI integration is to help a medical system already stretched thin, freeing up clinicians so they can spend more time with patients.

In some areas, like Vancouver’s St. Paul’s Hospital, AI is already in place, and it’s making significant improvements in efficiency. The WASP Lab is a tool at the hospital which analyzes culture plates to test for various bacteria or disease markers. There are a few components of the machine, including a robotic arm which loads the plates, and the AI integration that analyzes the cultures. “It’s very manual work. It’s very repetitive, and it’s relatively simple, but it’s not something we want a human to be doing a lot of; it’s not great for their wrists,” says Dr. Marc Romney, the head of microbiology and virology at St. Paul’s. Using this software frees up lab techs to do more critical thinking work. “We’re able to give (the lab techs) work that is safe, more intellectually stimulating, and it gives them a lot more time to do things that are more important.”

But don’t worry that your doctor will be replaced by an iPad. This tech still needs some supervision. Mr. Romney says they spent a few months training the AI to spot negative test results, and once it was performing that function with the same accuracy as a human, they moved on to other training, but they still audit the AI’s performance. So, the lab techs learned new skills, like trouble shooting and training, in order to work with the AI.

There are lots of areas in which AI might be helpful for doctors, like summarizing patient notes or reading up on the latest research. “There are 5,000 medical articles published every day in peer-reviewed journals. There’s no human being who could actually absorb all that information,” says Mike Fattori, vice-president of technology at Whitecap Canada, an AI driven consultancy firm. “You can take all of that information, and you can use it to train an AI model, and it will have all of that knowledge. In my mind, that’s spectacular.”

So, you don’t have to worry about AI taking over your job, but it’s a good idea to brush up on some skills, especially in the realm of information or data processing. “If your job is reliant on taking information and creating new information, or [synthesizing] existing sources, those are the kinds of jobs that AI will get very good at in the future,” says Rahul Krishnan, assistant professor of computer science and medicine at the University of Toronto. If you do a lot of data processing in your work, upskilling to learn compatible skills, like communication or visualization of that data could be a good plan.

It’s not a scary thing, but an opportunity to learn, says Mr. Popowich. “You might view it as working with a new colleague, but it’s a machine.”