AI will be in everything, everywhere…but not all at once

Artificial intelligence (AI) has the potential to become a global tool that redefines labour tasks, increases capital efficiency and drives productivity gains across multiple sectors, says Kathrin Forrest, equity investment specialist with Capital Group.

“It seems incredibly difficult to put a box around this, compared to something like smartphones, where we can triangulate the market with estimates around things like number of users, the replacement cycle and so forth,” she said.

Forrest said the speed and scale of AI adoption is unprecedented and shows no sign of slowing down.

“It is very likely that we’ll see a broadening set of applications of these new technologies. And that is likely to stretch beyond IT, communication services and related sectors, and into the broader economy. This is more along the lines of everything, everywhere — it just won’t happen all at once,” she said, paraphrasing the title of a recent mind-bending movie.

She said the total addressable market of AI is nearly incalculable — but that hasn’t stopped people from trying to peg it down. Estimates range from US$2 trillion to US$15 trillion by 2030.

“What I take from that is that people really don’t know,” she said. “What’s the value of being smarter at everything? It’s just really difficult to calculate.”

For investors, the convergence of several powerful trends spurring AI development spells opportunity.

“We have more data, we have more sophisticated instructions, we have more storage capacity, we have lower storage cost, we have more computing power to support speed. Lots of things coming together,” she said. “All of these things are important ingredients to building AI models.”

Forrest said there is a growing assumption that better models will emerge without further technological innovation, due to input gains like more data, parameters and computing power.

“The question then becomes, ‘What is the natural limit to doing all of this? To adding more inputs and making those models better?’ That question hasn’t really been answered,” she said.

So far, a limited number of mega-cap companies have first-mover advantages that are driving enormous capital expenditure and profit potential. Their success hinges on access to critical ingredients like powerful computer chips, proprietary data and development capabilities.

“Common denominators for companies really centre around those that are positioned around areas of tight capacity and bottlenecks,” she said.

Management execution will be increasingly important, she added.

“There are certainly opportunities for management teams that are forward thinking, thoughtful and proactive,” she said. “But the opposite is also true. There is potential disruption for incumbent companies whose leadership teams fall behind in implementing AI strategies, or in observing how their competitive landscape is changing, how customer preferences are changing or how regulation is changing around them.”

For investors, due diligence and agility will be key to company selection, she said.

“As we go about our day-to-day company analysis, we want to think about evolving competitive pressures, regulation, input availability — including labour, for that matter — customer preferences, and the speed and path with which companies adapt,” she said.

More broadly, investment opportunities will evolve as AI continues moving into new areas. Among the many implications of AI will be fundamental changes for a broad set of industries and sectors, including supply chain management, insurance underwriting, and utility grid-and-load management.

**

This article is part of the Soundbites program, sponsored by Canada Life. The article was written without sponsor input.