Technology can do a lot of impressive things, but there are certain skills that just can’t be automated, and those attributes will be in greater demand in the age of artificial intelligence (AI).

Despite widespread anxiety over the impacts AI will have on the labour market, Rotman associate professor and executive director of its Self-Development Laboratory, Maja Djikic, says that as AI performs more technical tasks, the technology will also create greater demand for certain quintessentially human traits.

“Obviously both are important, because AI can’t do all the knowledge work by itself, but these human leadership skills will really put you into a higher category of value,” Djikic says. “If you’re working with other humans, you’re going to need these skills, because they deal with complex human interactions — and given enough time, all human interactions become complex.”

No matter how much the technology develops in our lifetimes, Djikic says there are three broad categories of human skills that won’t be offloaded to AI.

1) Navigating non-verbal human interactions

Today’s most advanced AI tools are built on advanced automated systems known as large language models, or LLMs, which identify patterns in existing text to generate original content.

This form of “generative AI,” in other words, is wholly reliant on written and verbal communication, but those are not the limits of human interactions, leaving significant gaps in their ability to understand and respond to human emotion.

“They can't build trust, they can't motivate, they can't mediate, they can't negotiate, they can't resolve conflict and they can’t read the room,” Djikic says. “Being able to look around the table and rapidly perceive people’s possible motivations, intentions or emotional states — and then conduct themselves accordingly — is a skill that humans have been developing for millennia, and it’s vital for being a successful professional.” 

While AI can offer some helpful suggestions on how to read a room, mediate conflicts or engage with a colleague who is struggling with a personal crisis, Djikic says that it can’t make those assessments itself.

That is especially important, she says, given that people aren’t always forthcoming with their emotional state and often don’t fully understand their own motivations.

“AI can't read your tone of voice, it can't read your level of anxiety when you're talking about different subjects, it can't perceive presence in the way that we as humans can,” she says. “I as your leader, on the other hand, might have observed you interact with people, or seen you getting excited about new technologies, and I can actually ask relevant questions to help you discover motivations which you're unaware of.”  

2) Taking responsibility

AI tools can offer a lot of helpful suggestions, recommendations and solutions, but they can’t be held accountable for their own decision-making — at least not at the individual user level.

That, Djikic argues, will keep humans preferring the guidance of their fellow Homo-sapiens in a wide range of professional settings, from leadership to customer service and beyond.

“Humans inherently do not want to interact with a system in which a human decision maker is not taking responsibility,” she says. “We’re happy to interact with the technology; however, there has to be a person behind it who can make a decision, who can reverse a decision, who can take responsibility and offer an appropriate response to the issue at hand, and only humans can do that.”

Even as AI becomes more integrated into the decision-making process, Djikic suggests there will always be a limit to how much we as humans are willing to outsource to the technology. At the end of the day, she believes there will always be a need for real people to stand behind those decisions, take accountability and make adjustments as necessary.

3) Creating purpose

AI can, in some cases, help you get to where you want to go — or at least offer some helpful suggestions on how to accomplish a wide range of goals — but it can’t set the direction of travel, or identify when you’ve started to veer off course. That, Djikic says, is firmly within the domain of human leaders.

“As a psychologist, I think of purpose as our place on our developmental arc,” she explains. “At every point in that arc we’re at risk of falling a little to the side, and usually the way we can tell that we’ve started to fall off the rails of purpose and meaning is that our emotional systems give us cues that we’re not aligning.”

This moment-to-moment navigation requires a certain degree of self-reflection and an understanding of our own motivations, something an automated tool just can’t provide.

 “We will always prefer to work with self-reflective professionals and leaders who understand their purpose, who have emotional intelligence, who know how to build trust, who can motivate and mediate and negotiate and solve conflict and, in the end, take responsibility for the decisions they make,” Djikic says. “When I look at all of these things that AI can't do together, fundamentally, it cannot become a leader.”

Liked what you read? Subscribe to the Rotman Insights Hub to get even more personal development tips and tricks delivered straight to your inbox every other week.