This is the third chapter of a three-part annual report on the global asset management industry and the trends shaping its future. The full 2026 Global Asset Management Report: An Imperative for Growth is available as a PDF download.
The asset management industry has spent the past several years experimenting with artificial intelligence through pilots and productivity tools. The next phase is more consequential as agentic AI systems change how asset managers generate insight, serve clients, and run their platforms. The result is an emerging AI-first model.
The implications extend beyond efficiency. AI expands analytical capacity, lowers the cost of personalization, and allows firms to scale operations with fewer constraints. These shifts erode many traditional sources of advantage in asset management, from information edge to analytical differentiation to the sheer coverage that scale once bought—with implications that fundamentally change the basis of competition.
Yet the industry remains early in the transition. Asset managers trail banks and fintech firms in scaling AI across core processes. (See Exhibit 1.) Most are still focused on pilots and incremental productivity gains. That approach is no longer sufficient.
Real advantage will come from a strategic playbook built on three imperatives:
- Be bold. Stop optimizing at the margins. An order-of-magnitude AI advantage requires structural redesign, not incremental tools.
- Focus. Back a small number of transformative programs that can deliver clear P&L impact.
- Go deep. Set a top-down ambition, but rewire the business bottom-up. Lasting agentic advantage primarily requires changes to the operating model, talent, and processes that embed AI into how work gets done, not by data and technology alone.
Transforming the Value Chain
As AI becomes embedded in client, investment, trading, and operational workflows, its effects are beginning to appear across the asset-management value chain. (See Exhibit 2.) The following sections show how this is playing out.
Stay ahead with BCG insights on financial institutions
Client coverage and product customization
AI changes the economics of client coverage. Agentic systems can handle much of the administrative load around acquisition, onboarding, reporting, and routine client requests, with the potential to free capacity by 35% to 50%. But the more important shift is how that capacity is used.
With effective deployment, firms can deliver deeper service at scale, allowing relationship managers to spend less time on meeting preparation, requests for proposal, due diligence questionnaires, and servicing workflows, and more time on mandate design, interpretation, and trust-building. Algorithms can construct and manage portfolios tailored to client-specific objectives—such as tax optimization, sustainability preferences, or risk constraints—making customization more scalable than before.
The result is a different client model—one that can deliver stronger satisfaction and retention, faster expansion into adjacent mandates and segments, and broader coverage without a proportional increase in headcount.
Investment research and portfolio construction
AI can increase the scope and speed of investment work. It can help teams broaden coverage within their existing remit, expanding the opportunity set by assessing more companies, themes, signals, instruments, and trade expressions without diluting attention. Research cycles will speed up, allowing hypotheses to be tested and refined more quickly, while continuous data streams support ongoing reassessment of positions and risks.
These changes will lead to higher throughput and improved decision quality. Coverage diversifies, signals strengthen, and risk can be managed more dynamically. The result is better risk-adjusted performance. AI has the potential to improve Sharpe ratio for managers by 5% to 20%, though the gains will not be evenly distributed. As baseline analytical capabilities commoditize, firms that have stronger judgment, differentiated data, and superior portfolio construction and risk management will win.
Trading and execution
In liquid markets, AI can build on the electronic trading stack, optimizing routing, algorithm selection, timing, market impact management, and real-time control checks. As AI automates manual workflows—by 70% to 80%—human traders can focus more on execution strategy, exception handling, block trades, and other high-value decisions.
In less liquid or less electronic markets, AI will play a broader role by reducing friction across request-for-quote handling, negotiation support, mandate interpretation, documentation review, compliance checks, and transaction workflow management. For a typical multi-asset manager with significant equity exposure, this could affect roughly 30% to 40% of desk volume, where execution remains more manual.
Investment operations
Investment operations are evolving toward agentic systems that coordinate fund accounting, reconciliations, portfolio analytics, corporate actions, and reporting. In private markets, AI agents can process capital calls, interpret unstructured documents, and calculate and validate waterfall distributions autonomously. In public markets, they can reconcile net asset values across custodians, process corporate actions, and differentiate true exceptions for escalation.
These advances will shift investment operations from a linear support function to a scalable platform for growth. Firms can absorb greater AuM, product complexity, and customization without corresponding increases in headcount. Agentic workflows can increase capacity by 55% to 65% and reduce operational costs by around 40%. Human effort can be redeployed to designing the exception-handling logic that makes the autonomous layer smarter over time.
Rewriting the Operating and Talent Model
AI will enable an order-of-magnitude expansion in coverage and speed, more continuous decision making, and far greater operational scalability. Those developments will reshape competition in three ways.
First, judgment can move up a level. The edge will no longer come from producing analysis but from deciding what to do with it. In some cases, firms will generate alpha by going against AI-driven consensus. Leaders, however, will have to decide which models to use, how to combine them, and when to challenge them—and then translate those choices into better portfolio construction and performance. Firms that do this well will command a premium.
Second, distribution and relationships will become the primary battleground. Technology will create scale, but relationships, credibility, and fiduciary confidence will determine who captures value. Trust is paramount. Clients will choose firms they trust to interpret complexity, exercise judgment when the system is uncertain, and stand behind outcomes.
Third, mass customization will create new business model opportunities. As AI lowers the cost of tailoring portfolios, capabilities once reserved for the largest mandates can be extended far more broadly. Firms can deliver highly customized portfolios across public and private assets to address tax and liquidity needs and mandate constraints at a cost that works beyond the top tier of clients. For those that can scale personalization, these capabilities will open new segments, product forms, and revenue models.
Together, these shifts will change how teams are built. (See Exhibit 3.)
In investment, judgment can take the place of analysis as the primary source of value. Portfolio managers will be able to redirect 5% to 10% of time saved from analytics toward higher-order decisions. Analysts can move from data gathering and first-pass modeling toward management engagement and differentiated insight, with 50% to 65% of traditional junior-heavy analyst capacity redeployed. As much as 70% to 80% of standard execution flow—including prechecks, order management, and fill monitoring—will be able to run autonomously, allowing traders to shift from managing routine orders to shaping execution strategy, handling exceptions, and managing illiquid or stressed situations.
The distribution role can center increasingly on commercial judgment rather than service administration. Automation will redirect 35% to 50% of capacity toward AI-enhanced mandate design, client interpretation, and high-stakes dialogue. Coverage can scale more easily, and the human contribution will become relational and advisory.
In operations, automation can handle the execution. The function can shift to oversight, focused on exceptions, controls, regulatory interpretation, and continuous improvement.
In technology, support will give way to strategy. Technology teams will be charged with building and governing the AI, data, and control infrastructure. The emphasis will be on agentic architecture, orchestration, and resilience rather than maintenance.
Navigating to the AI-First State
Moving to an AI-first state requires reconceiving how the firm operates, not simply adding new tools to existing structures. Here’s where to focus.
Redesign the operating model for AI-native scale.
Update core processes end to end, reducing layers, decision paths, and redundant review loops. Rebuild workflows around agentic AI and cross-functional teams to solve problems holistically. Embed governance into the operating architecture from the outset to ensure auditability, model traceability, and clear accountability. Retrofitting control at scale is much more challenging than building it right from the beginning.
Establish AI agents and systems as a cohesive stack.
Prioritize scalable agentic architecture—including shared model environments, orchestration layers, modular compute infrastructure, and a unified data architecture. Invest heavily in agents that have the context, tooling, memory, and governance to automate workflows end to end. Reaching 80% automation is not enough to capture real gains. Design agents for continuous improvement through structured feedback loops and define a clear path toward recursive self-improvement as technology matures.
Build the skill premium.
Recruit bridge leaders who combine AI fluency with market expertise and systems thinking. Embed AI literacy into hiring, promotion, and performance frameworks from the C-suite down. Systematically upskill the workforce in human-agent collaboration so they know what models can and cannot do and how to prompt, verify, and improve them to continuously expand automation and value.
Hardwire transformation into governance and incentives.
Lead AI transformation from the top, with clear sponsorship and capital behind it. Tie incentives and accountability to measurable AI outcomes, including performance scorecards and compensation. Sequence the transformation by starting with high-impact, low-resistance areas that deliver visible savings and build confidence.
Agentic AI creates an order-of-magnitude expansion in what asset managers can cover, how fast they can move, and how efficiently they can scale. Capturing that requires structural redesign, focused capital allocation, and a talent model built for human-agent collaboration. The firms investing in all three now are redefining what it means to compete in this industry.
