The ascent of AI has been the primary narrative for equity investors in 2023, leading to a remarkable upswing in Big Tech stocks such as Apple, Microsoft and Nvidia. Yet, amid persistently high interest rates and stubborn inflation, many question if this tech resurgence is sustainable and whether the AI run has legs.

It is a question that Jeremy Maggs, host of Investec’s No Ordinary Wednesday put to Simon Lapthorne, a Senior Research Analyst for Investec Wealth & Investment UK, in the latest episode of the podcast.

The potential of AI

The debate around AI extends beyond its impact on the stock market, says Lapthorne. 

“The tech sector's 'hype cycle' often sees innovative technologies surge in anticipation before crashing when they fail to meet inflated expectations. However, AI's robust fundamentals set it apart. AI is not simply about tech; it is a general-purpose technology like the internet and the smartphone. Its application extends to all industries and business functions, enhancing productivity at an unparalleled scale. As such, it holds the potential to be a game changer for the global economy, adding an estimated 7% to global GDP by 2030.”

AI risks and challenges

While its potential as a wealth creator is staggering, AI brings with it notable challenges and risks.

“Large language models are trained on the public internet. They're designed to provide plausible next words, not necessarily accurate answers,” explains Lapthorne. Generative AI models can contain significant bias as they are trained on data from the internet, which may reflect historical biases.

“If you ask one of the new AI image generators to give you a picture of a certain type of person, you'll find the likelihood is that doctors are men, nurses are women, CEOs are white and investment analysts wear glasses,” says Lapthorne.

In addition to inaccuracy and bias, a potentially more dangerous risk is around authenticity, with deep fakes and voice cloning already making headlines. Questions around copyright infringement, fake news, privacy, security and authenticity have governments scrambling to try and regulate AI which could hamper its growth in the medium term, warns Lapthorne.

Picking the winners and losers in AI

Lapthorne says the short-term future for AI looks promising, but the question of discerning winners from losers in this fast-moving and unpredictable environment remains.

For AI investing, there's a risk-reward spectrum where the higher the exposure to AI, the higher the potential reward and risk.

“Companies like Microsoft, Alphabet, Meta and Amazon are very nicely exposed to AI, but it isn't all of their business. So if you invest in something that has an AI angle, but it isn't the be all and end all, that helps to de-risk it,” explains Lapthorne.

Companies in the AI supply chain like Nvidia, which makes the computer chips used in AI systems, can benefit regardless of which AI model dominates. Going even further down the supply chain can also help reduce risk. Lapthorne calls this approach “backing the picks and shovels merchants to the gold prospectors”: “You can go further down the supply chain to companies that make the machines, that make the chips, that go into the computers, that do the processing and the networking and peripheral areas if you're not sure who's going to win.”

AI in investments

While AI has yet to be fully harnessed for investment decision-making, its ability to augment productivity, combined with its far-reaching applications, make it a fundamental aspect of our future - one that will continue to shape and disrupt our investment markets for the near future.