Portfolio teams at large asset managers are approaching the fourth quarter with a cautious blend of conviction and hedging as froth concerns around artificial intelligence collide with conventional market drivers: earnings season, central bank policy and geopolitical tensions.

“We see AI investment as an antidote to weaker areas of the economy,” said Derek Hynes, a fixed-income portfolio manager at Wellington, noting that the theme has materially buoyed corporate spending and equity valuations. “But any loss of confidence in this area may have an outsized impact on financial conditions and so warrants careful monitoring.” His comments reflect a growing unease among allocators about the market’s dependence on a small cohort of AI leaders to sustain broad gains.

Momentum in AI-related equities remains strong, fund managers say, supported, in many cases, by accelerating revenue and profit upgrades. “It’s difficult to bet against the AI-related baskets,” Will Mcintosh-Whyte, a multi-asset fund manager at Rathbones, said. “Momentum remains strong, and continues to be, in the main, backed up by earnings.” That endorsement of fundamentals is tempered by reality: a handful of mega-cap tech companies have driven a disproportionate share of returns this year, leaving indices vulnerable if investor sentiment shifts.

Asset allocation moves vary. Equity managers are trimming extreme winners and rotating into more cyclical sectors that have lagged, seeking a valuation cushion. Some are locking in gains through covered-call overlays or buying put protection on concentrated positions. Fixed-income teams are managing duration carefully, mindful that any Fed signal toward tighter policy could push up discount rates and dent richly priced growth stocks. Cash buffers are being increased at some firms as a tool to exploit post-earnings or risk-driven dislocations.

Market data show the phenomenon: AI-themed ETFs and software and semiconductor stocks have seen heightened inflows and outsized performance versus the broader market, and are trading at premiums to S&P 500 multiples. Those premiums amplify sensitivity to higher discount rates and to unexpected downgrades in sales or margins tied to AI deployments.

Geopolitical frictions and regulatory scrutiny further complicate the backdrop. Export controls, supply-chain constraints for semiconductors and rising scrutiny on AI safety and competition policy could all act as catalysts for volatility. Managers emphasize that those non-economic shocks could be as consequential as earnings surprises in the months ahead.

Policy risk remains front and center. With central banks still data-dependent, a hotter-than-expected inflation print or resilient labor market could prompt officials to extend restrictive settings, pressuring long-duration growth assets. Conversely, signs of economic slowing would test the earnings-based case for many AI plays.

Longer term, most managers interviewed remain constructive on AI as a structural growth driver that will reshape industries over years, not quarters. But the consensus is also clear: near-term positioning must reconcile that secular upside with short-term fragility caused by concentration, valuation gaps and macro uncertainty. How managers balance those forces this quarter will likely determine whether markets continue to reward AI winners or punish them in a sudden repricing.