With U.S. stock indexes scaling new heights and AI chip giant Nvidia now worth more than $4 trillion, professional money managers are searching for ways to profit from the bull market while sidestepping the froth. Many are turning to a familiar playbook from the dotcom era: pivoting out of overheated stocks just before their peak and rotating into underpriced companies poised to benefit from the next wave of investor enthusiasm.
Francesco Sandrini, head of multi-asset and Italy CIO at Europe’s largest asset manager, Amundi, said his firm was adopting the same approach that worked during the late 1990s technology surge. “What we are doing is what worked from 1998 to 2000,” he noted, citing “irrational exuberance” in AI-related options trading but predicting the tech rally still had room to run. His focus, he said, was on “the highest growth opportunities that the market has yet to price in,” including software, robotics, and Asian technology companies.
The recalibration comes as investors begin to rotate out of the so-called Magnificent Seven—the elite group of U.S. tech stocks, including Nvidia, Microsoft, and Alphabet—whose collective market value now commands a historically large share of the S&P 500. While some fear missing out on AI’s explosive potential, others see danger in overstretched valuations.
Simon Edelsten, Chief Investment Officer at Goshawk Asset Management, compared today’s market to 1999, warning that “the odds of this AI boom being a bust are very high” as companies pour trillions into a market that “does not yet exist.” Yet, like many veterans of the dotcom era, he sees opportunities in the periphery—such as IT consultancies and Japanese robotics firms likely to benefit from AI infrastructure spending. “When someone strikes gold, you buy the local hardware store,” he quipped.
Historical precedent supports this cautious opportunism. A study by economists Markus Brunnermeier and Stefan Nagel found that hedge funds during the dotcom bubble largely avoided betting against overvalued tech stocks. Instead, they timed sector rotations adeptly enough to outperform the market by roughly 4.5% per quarter from 1998 to 2000, exiting high-flying names before retail investors piled in.
That same nimbleness is now being tested as AI mania spreads beyond semiconductors into energy, infrastructure, and industrial automation. Fidelity International’s Becky Qin is betting on uranium, arguing that AI’s massive power demands could drive up nuclear energy usage. Similarly, Kevin Thozet of Carmignac has been taking profits from top AI names to build exposure in Taiwan’s Gudeng Precision, which supplies critical equipment to chipmakers like TSMC.
Still, others caution that the parallels with the 1990s are not merely academic. Arun Sai, senior strategist at Pictet Asset Management, warned of “the building blocks of a bubble” already forming, noting that the rush to build AI data centers could mirror the overcapacity seen in telecom fiber optics two decades ago.
For investors like Janus Henderson’s Oliver Blackbourn, diversification remains the safest hedge. He is offsetting his U.S. tech exposure with European and healthcare assets, wary that any AI-driven correction could ripple across the broader economy. “We’re in 1999 until the bubble pops,” he said.
As artificial intelligence continues to reshape markets and industries, asset managers are walking a fine line—eager to stay in the game, but determined not to repeat the mistakes of the last great tech frenzy.