A September 2026 joint report by Morgan Stanley and Oliver Wyman found that only 8% of fund managers reported measurable improvements in investment returns from AI, despite growing adoption across asset management. The report, Breaking from the Pack: The Race for Above-Market Growth, cited the Mercer 2026 AI in Asset Management Survey. That survey found 55% of asset managers had integrated AI into at least one part of their investment process, while another 27% were testing applications.

Reported benefits were more common in operations and research than in returns: 69% of asset managers said AI improved operational efficiency, and 55% reported better insights and decision-making. Only 24% reported reduced costs. The report said faster research and productivity gains do not automatically produce financial gains across an organization.

AI use was most established in investment research, where 49% of firms had integrated it, followed by data processing at 40%. Adoption in core investment decisions remained lower: 13% had integrated AI into portfolio construction and 7% into trade execution. The report said it remains unproven whether broader research coverage will lead to persistent alpha, meaning returns above market benchmarks. It also said AI's ability to consistently improve investor returns remains uncertain.

A September 2026 joint report by Morgan Stanley and Oliver Wyman found that only 8% of fund managers reported measurable improvements in investment returns from AI, despite growing adoption across asset management. The report, Breaking from the Pack: The Race for Above-Market Growth, cited the Mercer 2026 AI in Asset Management Survey. That survey found 55% of asset managers had integrated AI into at least one part of their investment process, while another 27% were testing applications.

Reported benefits were more common in operations and research than in returns: 69% of asset managers said AI improved operational efficiency, and 55% reported better insights and decision-making. Only 24% reported reduced costs. The report said faster research and productivity gains do not automatically produce financial gains across an organization.

AI use was most established in investment research, where 49% of firms had integrated it, followed by data processing at 40%. Adoption in core investment decisions remained lower: 13% had integrated AI into portfolio construction and 7% into trade execution. The report said it remains unproven whether broader research coverage will lead to persistent alpha, meaning returns above market benchmarks. It also said AI's ability to consistently improve investor returns remains uncertain.