AI is incredible. But what exactly should we be doing with it?

Navigating volatility, economic pressure and accelerating technological change all at once is a challenge for even the greatest leader. Nowhere is that tension more visible than in the promises and pitfalls of artificial intelligence. The most forward-thinking organisations are already using AI to unlock value in a number of ways: faster decision-making, more inclusive design, bridging skill gaps and augmenting human capabilities. But in general, the technology is leagues ahead of the level of strategic thinking that’s required to make the most of it.

How and where to use AI, and what the benefits can be – now and in the future – is at the core of a recent Think Ahead panel event. Rebecca Dykema MBA2011, a tech and media business builder, and Matt Altass SEP109 (2023), Vice President of Strategic Growth at Array, joined Keyvan Vakili, Associate Professor of Strategy and Entrepreneurship and Academic Co-Director of the Data Science and AI initiative at LBS to explore both the complexities of AI and the vast opportunities it offers.

The event kicked off with an audience poll: What's the biggest barrier to businesses unlocking real value from AI? Interestingly, only 14% of the audience responders blamed a shortage of skilled talent. There are enough people with the tech skills, pretty much. It’s the other organisational capabilities that are dragging behind: 28% of respondents cited lack of clear business strategy and vision, while 23% blamed unclear ownership and fragmented approach across businesses.

“A lot of the challenges around AI are the good old-fashioned organizational challenges,” says Keyvan. “The culture, the infrastructure not being ready. ‘They push us but they don’t give us the money.’” This theme came up again and again: the tech is a tool, we humans have to work out what to do with it.

We are in a tools-first moment, and we need to move to strategy-first, and to think more expansively. One of the challenges is knowing where to start: businesses wrestle with which aspect of AI to deploy among the dazzling and yet troublingly vague possibilities that people are endlessly talking about.

“AI brings so much to the table,” says Rebecca. “You're seeing disenchantment from executives around a lack of return on investment, because they're really only scratching the surface of what its capabilities are. Having worked at the cutting edge this conversation with a number of different large brands, those that are really seeing impact are those that are thinking about their business goals and objectives, and looking at key pressure points and saying, how do we apply AI here in order to solve a problem?”

Matt agrees: “Unless you understand the job that you want the AI to do, you can't have a strategy for implementing, and you can’t have the training for your staff on using the AI.”

Despite some pieces of academic research demonstrating productivity gains from using AI in customer service and retail, most businesses are failing to crystallize the impact in terms of AI execution.

“Leadership has to be really focused on the challenge,” says Rebecca. “You need to be crystal clear on what your key goals and objectives are. What is your competitive differentiation, and how do you take technology and plug it in systematically to enable you to that strategy to life? That's the place to start.”

The challenge is that AI operates in narrow pockets of application, says Keyvan. And there is usually a gap between the people thinking about the strategy, who don't have visibility into processes, and the people who have that visibility and can surface opportunities for applying AI but are not in a position of power to make decisions.

Organisations are messy, and political, he points out, and it takes time for them to figure out how to implement anything new and scale it. Hearteningly, he is able to point to a lot of positive uses of AI, such as a company that used it to upskill novice developers to get their performance closer to the few who had the highest level of expertise.

“Companies can be proactive about using AI for good,” he says. Many organisations are now using the technology to de-bias recruitment practices, for example.

Then there are the obvious cost efficiencies. In the legal industry, says Matt, AI is able to produce 10 agreements in one day when it previously took a lawyer a week to produce one. However, take care, and check everything: “AI is brilliant. I couldn't function today without AI. But there are so many flaws!”

A case was thrown of court not long ago because a hapless lawyer had used AI and it had invented cases that had not existed. “AI can save you an enormous amount of time, building a skeleton of your contract, argument, and workflows. But you’ve got to know what you're doing. Don’t rely on the machine.”

It isn’t a given that lowering costs and increasing efficiency means job cuts, says Keyvan. To take a healthcare example: In Copenhagen, where almost 90% of mammograms to screen for breast cancer are carried out by AI, there is still a second reading by a human. So what next?

“You could say: well, you know, we don't really need as many radiologists. Or, you could say, look, this AI is helping us save lives at a much lower cost. So instead of doing the screening once a year for women over 50, let’s do it twice a year for women over 30.”

To harness AI for growth and enhance the resilience of your business:

• Focus. AI implementation needs to be rooted in strategic vision. Declutter and drill down on the handful of things you can do to drive impact with AI.

• See AI through the lens of augmentation: human plus artificial intelligence equals super intelligence.

• Layer in the people who can help manage the change - people with interpersonal skills, vision and critical thinking.

• Give people the time, slack and incentive to experiment. This is crucial.

• Think change management: work out who your stakeholders are, and how you can get them on board and aligned. Then the politics and culture won’t get in the way.

Watch the film here: