Asset Managers' AI Spending Paradox: Fearing Over-Investment While Pouring in More

Global asset managers are spending heavily on artificial intelligence—but many are not sure it is money well spent. A new study commissioned by Clearwater Analytics, titled "GenAI and the Data Divide," surveys 178 senior executives at asset management firms worldwide and reveals a striking investment paradox: the industry is pouring record sums into AI even as two-thirds of leaders worry they have already gone too far.

The findings, published in July 2026, offer the most detailed look yet at how the world's largest money managers are approaching generative AI—and why data quality, not budget size, may separate the winners from the rest.

The Spending Surge: AI Budgets Are Growing Fast—and Getting Faster

The study documents an extraordinary acceleration in AI investment. Among the surveyed firms:

• 63% increased AI spending by more than 50% in the past 12 months .
• 85% plan to raise their AI budget by at least 50% in the year ahead .
• 23% expect their AI budget to at least double in the coming year .
• No firm reported a budget decrease .

The shift is not new. More than half of firms (56%) began integrating AI into their operations four to five years ago, and only 9% started within the past year—meaning the industry is in the middle of its AI journey, not at the beginning .

The Investment Paradox: 66% Fear Over-Investment, 25% Want More

Perhaps the most striking finding is the lack of consensus on what the right level of AI spending actually looks like:

• 66% of fund managers fear their firms are over-investing in AI .
• 25% believe their organisations are still not investing enough .

This contradiction, which Clearwater calls the "investment paradox," points to a deeper uncertainty. As Souvik Das, Chief Technology Officer at Clearwater Analytics, noted, the industry is struggling to calibrate its commitment to a technology that is clearly transformative but whose return on investment remains unpredictable .

Where AI Is Expected to Deliver: Data, Decisions, and Risk Models

The study asked executives where AI will have the most impact in the next 12 months. The results point squarely at operations that involve processing information and supporting human judgment:

• 62% of fund managers expect transformative change in data generation and summarisation .
• 58% cited a major impact on decision-support systems, including portfolio rebalancing .
• 57% cited a major impact on predictive modelling and stress-testing .

These numbers suggest the industry sees AI not as a replacement for portfolio managers but as an accelerator for the data-heavy, analytical work that underpins investment decisions.

Data Quality: The Hidden Difference Between Success and Stagnation

Not all AI spending produces the same results. The study found that the quality of a firm's underlying data is a decisive factor in whether AI investments actually improve performance.

• 44% of firms with good or excellent data accuracy say risk management has become "much more proactive" as a result of AI .
• Only 17% of firms with moderate or poor data can say the same .

"Asset managers are sharply increasing their spending on AI, but bigger budgets aren't always translating into results," Institutional Investor reported, summarising the study's core message . Firms with stronger data governance are extracting meaningful operational improvements; those without are spending money without seeing comparable gains.

What This Means for the Industry

The "GenAI and the Data Divide" study paints a picture of an industry making bold bets on a technology whose payoff is still uncertain. Budgets are rising without clear benchmarks for success. The majority of firms are already several years into AI integration, yet two-thirds of their leaders think the organisation may be spending too much.

The clearest actionable finding is that data quality matters more than budget size. Firms that invest in accurate, well-governed data are far more likely to see AI improve their risk management and decision-making. For asset managers trying to navigate the AI landscape, the study suggests that the priority should not be how much to spend—but how clean the data feeding the models actually is.