Recommended Free Tools
Wealth managers are spending more on AI because it can speed up research, analysis, and routine workflows—but many firms still lack the data foundations and measurement systems needed to show what that spending returns. Available surveys point to reported efficiency gains, not a proven industry-wide lift in investment performance. That distinction explains how AI use can expand while its return on investment remains difficult to establish.
Why AI budgets are growing faster than proof of returns
AI can help employees process information, draft or summarize material, and support analysis. Those benefits may be useful even when they do not immediately translate into measurable profit or better portfolio results. Establishing a financial return requires firms to define a baseline, account for implementation and oversight costs, and connect a change in workflow to a business outcome.
As an Amazon Associate I earn from qualifying purchases.
InvestmentNews reported in July 2026 that most firms in an F2 Strategy survey had not established formal methods for measuring AI project returns; none of the bank and trust respondents had done so. The article describes the survey population in two ways: it refers to firms with $31 trillion in assets under management, while also saying underlying data came from 40 leading RIAs, wealth-management firms, and broker-dealers representing $8.6 trillion. Those figures describe different reported aspects of the survey and should not be treated as the same sample denominator. InvestmentNews’ account of the F2 Strategy survey also reports that 64% of wealth-management firms surveyed lacked a unified data layer, as did 83% of bank and trust respondents.
F2 Strategy’s reported results show why “elusive” does not mean “no value.” Among firms measuring AI investment, 68% reported at least 25% greater efficiency in targeted workflows. That is a survey-reported result for specific workflows—not a 25% increase in firmwide productivity, revenue, or profits.
#1 Best Overall
What the surveys say—and why their percentages differ
Adoption statistics are not a single industry benchmark. The surveys below cover different geographies, firm populations, time periods, and definitions of AI use. Some measure current deployment, others include planned use or explore perceived impact. Their percentages should be read within those limits rather than combined.
| Source and population | Reported finding | What it measures |
|---|---|---|
| FCA, 2026; UK discretionary portfolio-management firms | 13% used in-house or third-party AI tools; 45% when firms considering use in the following 12 months were included | Use at the time of collection, then use plus near-term consideration. The FCA says adoption may since have risen. |
| EY, 2025; 100 wealth and asset managers | 95% had scaled GenAI to multiple use cases; 27% of all respondents reported substantial GenAI impact over the past 1–2 years | Deployment breadth versus respondents’ reported business impact. EY separately says 78% were exploring agentic AI. |
| Mercer, 2026; global survey of 131 asset managers conducted in February 2026 | 55% had integrated AI into at least one investment process; 91% planned to increase AI use in the next 12 months | Current integration and stated plans, not achieved returns. |
| DSIT AI Adoption Survey, as reported in the UK Financial Services AI Adoption Plan, 2026 | 21% of firms in financial and real estate sectors had adopted AI in early 2025, versus 16% across the economy | Sector adoption in a broader survey. The plan also discusses a distinct FCA/Bank of England survey, published in 2024, with adoption around 75%. |
The UK figures illustrate why the denominator and timing matter. The FCA’s 13% figure is a snapshot of surveyed wealth managers’ reported use; its 45% figure adds firms considering use in the next year. The government plan’s 21% covers financial and real estate firms, while its roughly 75% figure comes from another survey and a different measure. None is a direct substitute for another.
EY’s survey likewise separates scale from impact: many respondents reported multiple GenAI use cases, but only 27% of the full sample said GenAI had delivered substantial impact in the prior one to two years. EY described early gains as concentrated in compliance, risk management, and IT, with sales and marketing, client services, acquisition, and onboarding emerging as areas for savings. This is a consulting-firm survey, not a census of all wealth managers. EY’s wealth and asset management findings also identify concerns about regulation, privacy, inaccurate outputs, hallucinations, and bias.
Efficiency is easier to report than investment outperformance
AI-supported work can be assessed through time saved, processing speed, or the quality of information and analysis. Establishing that AI caused stronger investment returns is harder: markets change, portfolios face different risks, and many other decisions affect performance.
Rank #3
- Ideal for Gifting
- Ideal for a bookworm
- Compact for travelling
Mercer’s February 2026 survey makes the distinction visible. Respondents reported using AI for operational efficiency within existing teams (73%) and as a partner for investment-process insights and analysis (68%). Only 5% said they gave AI autonomous or semi-autonomous authority over investment recommendations or trades. On outcomes, 69% cited enhanced operational efficiency and 55% faster or higher-quality insights, while 8% reported measurable improvement in investment returns and 8% reduced portfolio volatility. These are respondents’ reports, not evidence that AI caused those results. Mercer’s Global Manager Research Leader Beverley Sharp characterized the technology as “largely a partner rather than a decision-maker.” Mercer’s 2026 asset-manager survey describes the findings.
The pattern is consistent with an augmentation-first approach: AI supports employees rather than independently setting recommendations or executing trades. Operational gains can justify a use case on their own, but they should not be presented as proof of better investment performance or improved client outcomes.
Rank #4
What makes AI implementation costly and difficult to scale
A useful AI system depends on accessible, reliable data and the ability to integrate with existing processes. Firms may also need to invest in security, validation, governance, and human review. If these costs are not included alongside workflow benefits, a project can appear successful operationally while its net financial return remains unknown.
Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →The Bank of Canada’s 2026 Financial System Survey offers a broader financial-sector view, not a wealth-manager-only estimate. Respondents cited difficulty integrating AI with existing infrastructure and workflows (58%), talent constraints (56%), data-security and privacy concerns (33%), and high implementation and use costs (31%). The survey also describes firms using AI to finish tasks faster and redirect staff to higher-value work, while noting that some had not quantified the benefits and that infrastructure, governance, validation, and oversight costs complicated ROI assessment. The Bank of Canada’s Spring 2026 survey includes planned uses such as investment research and management, risk monitoring, operations, financial-crime prevention, and customer service.
- Data quality and integration: Fragmented or unreliable data can limit what a tool can do and make results difficult to compare.
- Privacy and security: Sensitive financial and client information raises requirements for access controls and handling.
- Accuracy and explainability: Incorrect outputs, hallucinations, and bias can undermine trust, especially when decisions affect clients or portfolios.
- People and process: Skills, staff training, workflow redesign, and ongoing human review carry costs beyond the initial technology purchase.
- Governance and compliance: Firms need clear responsibility for system use, validation, monitoring, escalation, and client impact.
How firms can measure AI ROI more credibly
There is no single industry-wide AI ROI figure established by these surveys. A firm evaluating a project should compare like with like: define the workflow and baseline, include the full cost of operating the system, and measure the intended outcome at the same level where the tool is used.
- Choose a bounded use case. Specify the task, users, data, and whether AI assists a person or has authority to act. Keep workflow support distinct from client-facing advice or investment decisions.
- Record the baseline. Before deployment, measure current time per task, throughput, error or rework rates, service levels, and relevant quality indicators.
- Count full costs. Include integration, data preparation, licenses or usage, staff time, security, validation, governance, and ongoing oversight.
- Compare after deployment. Track the same measures over a defined period and account for changes in workload, staffing, or process that could explain the difference.
- Separate operational and client outcomes. Time saved or faster analysis is not the same as better service, fair value, or improved investment results. Measure each claim with an appropriate outcome.
- Set review and stop conditions. Assign an owner, monitor accuracy and exceptions, and pause or revise use if quality, privacy, or client-impact thresholds are not met.
For investment performance, firms should be especially cautious about attributing returns or volatility changes to AI. The available Mercer results are self-reported and do not establish a causal effect; the other cited adoption surveys likewise do not provide an industry-wide causal estimate.
Why governance matters to clients, not just compliance teams
AI use in wealth management can affect recommendations, service, financial-crime controls, and how client information is handled. Governance therefore belongs in the business case: it shapes whether a workflow can be deployed safely and whether its claimed benefits are acceptable for clients.
The FCA’s 2026 survey covers around 400 UK wealth-management firms, whose supervised portfolio supports more than 5.5 million retail clients and nearly £1 trillion in assets. The regulator places responsible technology use alongside governance, financial-crime controls, fair value, and effective client support. FCA Director of Consumer Investments Lucy Castledine said firms need “clear governance, strong financial crime controls” and should provide fair value and effective client support as they use technology, including AI. These points describe the FCA’s UK context, not a universal legal checklist. The FCA’s 2026 wealth-management survey also notes that its AI adoption figures reflect submissions at collection time.
The UK government’s 2026 Financial Services AI Adoption Plan records industry requests for practical guidance on applying existing principles to Consumer Duty, model risk, explainability, and accountability. That is a description of the UK policy conversation; obligations and regulatory approaches differ by jurisdiction. The UK Financial Services AI Adoption Plan discusses both potential productivity and client benefits and the need to address implementation concerns.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
