How should investors position for the robot apocalypse?

I find myself having some odd conversations with fund managers these days. After years of asking them about things like interest rate differentials and allocations to cash, now the questions cropping up include “what do you think Donald Trump means when he threatens military action on the bond market?” and “how soon do you think Elon Musk will establish a functioning colony on Mars?”

So in a way, it was the new normal this week when the starting question to many chats was “how are you positioning your portfolio for the robot apocalypse?”

It is a ridiculous and yet serious question given the apparently sincere warnings from big tech heavyweights over the past week that — oopsie — the AI widgets they have created and that have made them fabulously wealthy could soon seek to move on from semi-autonomously hacking websites and start killing humans at scale.

This seems bad. It is a bad sign for the corporate governance and basic management practices inside large unlisted tech firms, and moreover, it’s pretty bad for humans. I have never previously given this much thought, but presumably, given that continuing human existence is an important foundation for consumer spending and resource utilisation, it is also bad for financial markets. This feels like a testing moment for stock markets that are highly concentrated in tech in general and AI in particular.

It is striking, then, that the market reaction to all this has been a shrug. Stocks did start the week with a wobble after all these weekend warnings, particularly in the tech sector. The famously AI-infused Korean stock market shed more than 3 per cent. But crack out a chart of the US’s big-hitter S&P 500 stocks index or even the techy Nasdaq, and I challenge you to identify the spot where some of the richest and most influential men on Earth revealed that they may accidentally have created sentient weapons of mass destruction.

The resilience in stocks is only half of the picture. Over in the haven side of the market, such as government bonds, it is also tough to see any scar. Typically, US government bonds and the dollar respond favourably when bad news lands — a pattern that still largely holds despite the best efforts of the US administration to chip away at their reserve asset status. This time around, however, bonds remained at their weakest levels in decades. It seems investors are more afraid of fiscal deficits than of demonic robots. Gold, too, remained stable — no sign of a fear trade there.

Sometimes non-reactions tell us as much about the market regime as reactions. So what is this all about?

The first possible reason for this insouciance is that in a warped kind of way, these warnings might be good for stocks in the AI ecosystem, not bad. The hyperscaler companies rushing to build out AI capabilities are turning to the corporate bond markets for the billions they require. They are not doing this out of distress, as such, but nevertheless, a slightly more sober pace of growth in spending on AI could, in fact, be helpful.

One investor also suggested to me, only half joking, that a new industry might now spring up, seeking to defend us from destructive AI agents. A new investment opportunity awaits, perhaps.

Another possible explanation is that investors simply do not believe the frontier AI lab leaders on this. Instead, money managers see it as either a weird kind of marketing, where OpenAI and Anthropic duke it out to claim theirs are the most dangerous and sophisticated models, or a contorted excuse to take time out and figure out how to make these models make money. Neither reflects well on an investment industry that will almost certainly bestow stratospheric valuations on those two companies when they eventually list on stock markets, even with the current leadership that has, by its own admission, failed to keep its technology under control.

Recommended

The final reason for calm is that maybe these warnings really do reflect a genuine danger, but it is a risk that is impossible to hedge. No safe asset will retain its value if we do all get zapped by malign, recursively self-improving AI tools and honestly, we are unlikely to be checking our pension portfolios and 401(k)s at that point anyway.

The answer, then, is to just keep on going. Stock markets have either ignored or bounced back from a huge range of shocks just since the pandemic — the kiboshing of geopolitical alliances, two wars involving key energy producing nations, a US regional banking crisis, you name it.

Underpinning it all is the assumption that if risky assets were to tumble in value, for any reason, central banks and governments would ride to the rescue as they have done so consistently since the financial crisis of 2008.

It is all rather uncomfortably nihilistic, but fair. And no one wants to be the fund manager who jumps out of stocks for fear of a robot apocalypse that, honestly, could be the figment of AI engineers’ imaginations, only to land in bonds that get chewed up by fiscal incontinence that we know to be real. While the AI-generated music is still playing, you have got to get up and dance.

[email protected]