AI Reshapes Asset Management: Profit Margins Could Rise to 44.6% by 2030

Overview

This research report quantifies the impact of generative AI (Gen AI) on the global asset management industry through 2030. Bernstein adopts an “early‑action” stance, viewing AI implementation as an urgent priority whose benefits will begin to materialize significantly from 2028 onward. The report projects that, as AI adoption follows an S‑curve trajectory, industry spending on AI will peak in 2028 and then stabilize. Meanwhile, efficiency gains are expected to substantially boost sector profit margins, reducing the cost‑to‑income ratio from a long‑term stagnation of 68.5% to around 56% by 2030. Beyond serving as a technological upgrade, AI is also seen as a critical survival mechanism in the face of fee compression driven by the rise of passive investing.

Core views

AI Spending and Adoption Rate Projections: The research report forecasts that global assets under management (AUM) will reach approximately $200 trillion by 2030. Based on the Bass diffusion model, AI adoption is expected to rise from 55% in 2026 to 88.2% by 2030. Accordingly, the AUM addressable by generative AI is projected to grow from $51 trillion in 2025 to $176 trillion in 2030, at a CAGR of 28%. Against this backdrop, total industry spending on AI is anticipated to increase from $9 billion in 2025 to $32.5 billion in 2030, at a compound annual growth rate of 29%. Notably, 2028 is set to be the peak year for AI expenditure, at roughly $36 billion; thereafter, as costs associated with in‑house development are amortized and maintenance models mature, spending intensity will gradually decline and stabilize. Diverging Efficiency Gains and the Inflection Point: The report highlights that the efficiency gains delivered by AI are not evenly distributed but exhibit clear tiered differentiation. AI leaders—large institutions representing about 20% of industry AUM and already deploying production‑ready agentic AI—are expected to achieve an 8% efficiency uplift by 2026, rising to 26% by 2030. In contrast, followers (Scalers), accounting for 45% of AUM, and laggards, comprising 35% of AUM, start later, posting efficiency gains of 4% and 0%, respectively, in 2026, which climb to 22% and 17% by 2030. The industry‑weighted average efficiency gain is projected to increase from 0% in 2025 to 21% in 2030, with long‑term steady‑state levels reaching 34%. Among functional areas, middle‑ and back‑office operations, along with technology functions, offer the greatest potential for efficiency improvement (40%), followed by central functions such as compliance and HR (35%), and front‑office investment research (30%). The year 2028 marks a critical inflection point, when AI‑driven efficiency gains (12%) are expected to exceed incremental AI costs (6.8%) for the first time. Offsetting Margin Expansion and Fee Compression: Over the past decade, the asset management industry’s cost‑income ratio has hovered around 68.5%, under persistent pressure from fee compression. The report argues that AI will serve as the sector’s “savior.” Quantitative analysis suggests that AI‑enabled cost savings could total approximately $200 billion by 2030, fully offsetting the adverse impact of declining fees. As a result, the industry’s overall operating margin is projected to improve from 32.4% in 2026 to 44.6% in 2030, with the cost‑income ratio falling to 56%. For AI leaders, margin expansion will be even more pronounced, climbing from 35% in 2026 to 48% by 2030. The report underscores that large firms, leveraging their infrastructure advantages, and boutique shops, capitalizing on their agility, stand to benefit most, while mid‑size players may face margin pressure due to operational complexity.

Analysis framework

The research report employs a top-down modeling approach, complemented by bottom-up case validation. First, in estimating expenditures, it adopts a two-tier model based on “addressable AUM × spending intensity.” The first tier uses an S-curve (Bass Diffusion Model) to project AI adoption rates and, combined with PwC’s global AUM forecasts, derives the addressable market size. The second tier, informed by survey data, specifies deployment ratios for “Buy,” “Build,” and “Hybrid” models, along with their respective cost profiles—higher upfront costs with Build, stable and sustained expenses with Buy—and calculates AI spending intensity per million dollars of AUM, thereby arriving at total industry expenditure. Second, in assessing efficiency and profitability, the report disaggregates asset management firms’ cost structure into five major segments: front‑office (research/PM), middle‑back office, IT & data, distribution, and central functions. It then assigns different AI‑driven efficiency‑improvement assumptions to each segment—40% for the middle‑back office, 30% for the front‑office—before computing a weighted average of industry‑wide efficiency gains. These gains are translated into cost savings, which in turn inform the impact on operating margins and the cost‑income ratio. Additionally, the analysis introduces three typologies—AI leaders, followers, and laggards—to conduct sensitivity testing and highlight the financial‑performance divergence across varying adoption paces. Finally, by comparing the magnitudes of “fee compression headwinds” and “AI‑driven cost savings,” the report underscores the necessity of AI as a hedging tool, concluding that by 2030, AI‑enabled cost reductions will be 6.5 times greater than fee‑related losses.

Methodology notes

• Industry/Industrial Analysis FrameworkPenetration Rate S-Curve

  The S-curve model in the technology adoption lifecycle (Bass Diffusion Model)

  The research report employs an S-curve to model the penetration of generative AI in the asset management industry. This framework posits that the adoption of new technologies begins slowly, accelerates over time, and eventually reaches a saturation point. The report calibrates the 2026 adoption rate at 55%, using this projection to estimate the addressable market size over the coming years—a standard approach for assessing the market potential of emerging technologies.

• Company Fundamentals and Financial FrameworkOperating/Financial Leverage Analysis

  Cost-to-Income Ratio and Operational Profit Margin Analysis

  The core logic of the research report lies in analyzing how AI can boost profit margins by reducing operating costs (the denominator). By disaggregating the cost structure—covering front‑office, mid‑office, and back‑office functions—and applying distinct efficiency‑enhancement multipliers, the report quantifies AI’s cost‑saving effects on both fixed and variable expenses, thereby deriving the extent to which these savings translate into improvements in key profitability metrics, such as operating margin and the cost‑to‑income ratio.

• Competitive and Strategic FrameworkMoat / competitive advantage

  The Manifestation of First-Mover Advantage and Scale Effects in the AI Era

  The research report distinguishes among “AI leaders,” “followers,” and “laggards,” noting that large institutions that invested early in AI will establish data‑driven and process‑based barriers, thereby achieving greater efficiency gains and higher profit margins. This underscores how, during periods of technological disruption, first‑mover advantages and economies of scale can translate into sustained competitive edge, leading to divergent performance across the industry.

Key data

• Total Industry AI Spending in 2030US$32.5 billionIn 2025, it is projected to reach US$9.0 billion, with a CAGR of 29%, and peak at US$36.0 billion by 2028.
• Overall Industry Efficiency Gains by 203021%Under the long-term steady-state scenario, it could reach 34%, with the greatest improvement observed in middle- and back-office functions (40%).
• Industry Operating Profit Margin in 203044.6%In 2026, the figure is projected at 32.4%, with AI leaders potentially reaching as high as 48%.
• Industry Cost-to-Revenue Ratio in 203056%Over the past decade, the average stood at 68.5%, with a deceleration expected to accelerate after 2028.
• Total AI-Driven Cost Savings by 2030Approximately USD 200 billionSufficient to offset the negative impact of consumption rate compression.

Impact & implications

The research report argues that AI is not merely a technological tool but a strategic imperative for the asset management industry to address structural challenges—such as the downward pressure on fees driven by the rise of passive investing. For investors, this means focusing on large asset managers with first-mover advantages in AI infrastructure and data governance, as well as smaller boutique firms that can flexibly integrate AI‑enabled workflows. Mid-sized players that fail to effectively harness AI risk being squeezed out of the market. Moreover, with the emergence of AI‑native portfolio managers, investment processes are shifting from “human‑centric” to “AI‑driven, human‑governed,” with analysts evolving into research directors and AI overseers, while portfolio managers increasingly assume the role of AI orchestrators focused on risk management and conviction‑building. This transformation will blur the lines between quantitative and fundamental investing and may heighten systemic risks arising from algorithmic convergence.

Risks

• The actual efficiency gains from AI implementation may fall short of expectations, resulting in an ROI that underperforms model forecasts.
• Algorithmic herding may increase systemic risk in the market.
• Changes in regulatory policies may constrain the scope of AI’s application in financial decision-making.
• Data privacy and security concerns may impede the training and deployment of AI models.
• Talent shortages may lead to delays or failures in the implementation of AI projects.

What to watch

• Specific AI budgets and ROI data disclosed by major asset management firms.
• Real-world implementation cases and cost-saving outcomes of AI in mid- and back-office automation, such as compliance and KYC.
• The pace of industry fee compression versus the rate of AI-driven cost savings.
• Regulatory guidance on the compliance of AI-generated investment advice.
• Market acceptance and capital inflows for AI-native investment products.