'Investors should back the AI maximalists'

Britain’s investment establishment is firmly convinced that the US market – and in particular the technology-centric “Magnificent Seven” stocks – are an investment bubble that is destined to implode as the “hype” surrounding AI dissolves. Terrified by the speed and scale of the bull market, which they failed to foresee, they are desperately waiting for a repeat of the 2000-2003 collapse.

In the US, they think differently. The Magnificent Seven’s market share of the S&P 500 has doubled to 32% since the end of 2022, while its share of corporate earnings has risen from 13% to 23%, points out veteran strategist Ed Yardeni. The share of the information technology and communication services sector in the S&P index has just passed the peak of 40% reached in early 2000, but it now accounts for 37% of forward earnings against a peak of 24% in early 2000. So long as the earnings growth of the S&P 500 and the tech sector continues, the doomsters on this side of the Atlantic will keep being disappointed.

The AI maximalists

The bearish UK consensus makes the bullishness of the team managing the £5 billion Polar Capital Technology Trust (LSE: PCT) almost contrarian. Ben Rogoff, the lead manager, describes the team as “AI maximalists” who believe that AI will change the world and have focused the portfolio accordingly. Since April, AI winners have outperformed AI losers by 50%, he says. This accounts for the 36% return of the trust, compared with 27% for its tech sector benchmark. This followed a dull year in which the trust returned just 3% compared with a benchmark return of 5%.

Try 6 free issues of MoneyWeek today

Get unparalleled financial insight, analysis and expert opinion you can profit from.

Start your trial

Sign up to Money Morning

Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter

Don't miss the latest investment and personal finances news, market analysis, plus money-saving tips with our free twice-daily newsletter

Sign up

That said, Rogoff warns investors to expect volatility: “it’s pretty normal in a time of change”. During the 1995-1998 bull market, there were seven setbacks of more than 15%.

“You’ll only hear negative things about AI,” he says, “but we refute very strongly that we are in an AI bubble, rather a transformative technology era.” In periods of transformative change, the pace of improvement is very rapid. The invention of lifts by Otis in 1885 enabled the construction of skyscrapers and a trebling in the height of buildings to 42 metres. The Empire State Building, which was completed 46 years later, multiplied this nine-fold to 381 metres.

Explosive growth in AI

Global annual spending on information technology was $5.4 trillion last year, but the global wage bill was $44 trillion. “Therein lies the market opportunity.” Growth in AI has been explosive as it quickly became a corporate imperative: there are 700 million weekly users of OpenAI and one billion for Meta AI. Investors were briefly taken aback by China’s DeepSeek bursting onto the market, but Rogoff sees it as just a milestone on the path of progress with the cost likely to collapse. “This is a feature of the tech model, not a bug.”

The capital expenditure by the largest tech companies continues to rise, with compound annual growth of 40% expected between 2023 and 2027. “If DeepSeek was a problem, they wouldn’t be spending this,” says Rogoff. Yet AI capex still accounts for under 1% of US GDP. By comparison, “railroad capex reached 3% in the 19th century”.

While AI is an opportunity for some, it is a threat to others. Intel, the best semiconductor company in its time, started to lose investors’ confidence in 2015 “when they realised it wouldn’t be the conduit to the world of cloud storage”. Now software firms Adobe and Chegg are AI casualties because “the market is a discounting mechanism and so looks well ahead. AI is scaling faster than Moore’s law”. (Moore’s law is the 1975 prediction by the cofounder of Intel that the number of transistors on a microchip would double every two years – a trend for the growth of computing power.)

A broad portfolio with an AI focus

The trust’s portfolio has 90-110 stocks, “almost all explained by our AI focus”, but “we are not afraid to have zero weightings in firms in our benchmark”. A broad portfolio enables the managers to have benchmark or underweight positions in firms where they do not have a strong conviction, such as Oracle (up nearly 40% in a day recently) and Palantir (very pricey, “but you have to look at the scale of the opportunity”). Quantum computing is an “exciting technology, but despite significant breakthroughs in the last two years, it is still pre-revenue and at least five years away from profitability”.

Inevitably, nearly 75% of the portfolio is US listed, with less than 1% in the UK, while 97% is in firms with a market valuation above $10 billion. “We have twice the growth of the index for a limited valuation premium.” A third of the portfolio is in the Magnificent Seven, “but we expect that to come down”. Avoid a holding at your peril.

This article was first published in MoneyWeek's magazine. Enjoy exclusive early access to news, opinion and analysis from our team of financial experts with a MoneyWeek subscription.