International | 10:30 AM
Every new technology triggers scary 'what if' narratives, but investors should focus on the underlying dynamics of today's AI buildout, argues Jordan Stuart, Vice President, Investment Director at Federated Hermes.
By Jordan Stuart, Vice President, Investment Director at Federated Hermes
It’s likely that fear has accompanied every major technological revolution. In Victorian England there was a widespread fear of “railway madness.”
The idea was that the motion of the carriage could unhinge a person and cause insanity. Medical journals studied the subject, and the media was only too happy to promote the worry.
Investors should be careful not to confuse technological disruption with technological destruction.
Today, we see a similar dynamic at work with artificial intelligence. Concerns around job displacement, misinformation, safety, and energy consumption have moved onto talk of existential risk.
Perhaps this is understandable: It’s a new technology, and all such concerns deserve serious consideration. Investors should be careful, though, not to confuse technological disruption with technological destruction.
We won’t be destroyed
The most jarring headlines lately speak of researchers’ claims that AI could go rogue or bring on mass destruction.
We view such concerns as being wildly overstated, however. First of all, AI is more like a calculator, albeit a very useful one, than a human being. It lacks desire, and it has no goals except the ones we give it.
A recent widely reported incident involved AI agents hacking, which prompted fears of AI going rogue.
The truth is a bit more mundane. Humans could have stopped the hacking multiple times during the tests.
Also, the agents would eventually run out of budget, which many did.
Disruption, yes, but not depression
Beyond “going rogue,” the other main AI worry has to do with massive unemployment.
History repeatedly demonstrates that transformative technologies initially threaten established ways of working before ultimately expanding productivity, creating new industries and improving living standards.
The question for investors is therefore not whether AI will create disruption — it will — but whether that disruption ultimately destroys economic value or creates substantially more of it.
Consider agricultural mechanisation.
In 1900, 10.9 million American agricultural workers produced food for 76 million people. By 1950, only 7.5 million were feeding a population of 151 million.
The tractor and mechanised farm equipment dramatically reduced the labour needed to produce food, but America did not experience permanent mass unemployment.
Labour migrated into manufacturing, services, medicine, finance, technology and entirely new industries. Similar fears accompanied the Industrial Revolution, electricity, computers, ATMs and the internet.
Technological progress has rarely been painless, but over time it has allowed humanity to produce more with less.
Global life expectancy, for example, increased from roughly 32 years in 1900 to more than 70 today, in large part because of improvements in medicine, sanitation, nutrition, productivity and technology.
No bubble in sight either
Like electricity or the railroads, AI looks to be a general-purpose technology of world-changing magnitude. The final worry many investors have is that we’re in a bubble.
The physical infrastructure required to support it is still struggling to catch up with demand. Current industry spending on AI infrastructure is running about 2.5% of US gross domestic product, about the same as that spent on the railways in the 1870s.
At the same time, scarcity remains visible throughout the ecosystem: AI cloud providers are raising prices for access to high-end computing, chip suppliers continue to report demand exceeding available supply, and billions of dollars are being raised to construct additional AI computing capacity.
The opportunity consequently extends well beyond chips themselves to networking, memory, power generation, transmission, transformers, cooling, data centres and semiconductor equipment.
In short, this looks less like an industry with excess physical capacity than one racing to construct the infrastructure necessary for a rapidly expanding installed base of AI applications.
For investors, this helps make the case for keeping calm and maintaining a carefully considered allocation to what promises to be the defining technological leap forward of our age.
Re-published with permission. Views expressed are not by association FNArena’s.
Find out why FNArena subscribers like the service so much: “Your Feedback (Thank You)” – Warning this story contains unashamedly positive feedback on the service provided.
FNArena is proud about its track record and past achievements: Ten Years On