Meeting notes to CRM
With client consent, AI transcribes the meeting, summarizes it, extracts action items, drafts the follow-up email and writes the note back to the CRM. Morgan Stanley's Debrief tool works this way and saves notes into Salesforce.
AI in wealth management: use cases, data and regulation
Advisors spend most of their day away from clients, more than a third of them plan to retire within a decade, and most wealthy clients expect their firm to use AI. Here is what the latest data says, where AI already takes work off advisors and what regulators have said about it.
- 59% of advisor time goes to administration, compliance and other non-client work
- 73% of employee advisors and 42% of independents now actively use AI tools
- Summarisation and information extraction is the top generative AI use at FINRA member firms
The advisor capacity problem in numbers
- 59%of advisor time goes to administration, compliance and other non-client tasks
- 37.4%of US advisors plan to retire within 10 years, holding 41.4% of client assets
- 98%of Morgan Stanley financial advisor teams adopted its generative AI assistant
- 39.2%of US finance and insurance firms used AI in their operations in October 2026
AI use in finance and insurance rose about 9 points in a year
The Census Bureau asks US employers every two weeks whether they used AI in any business function. In finance and insurance, a sector that includes banks, insurers and investment firms, the share rose from about 30% to about 40%, and 43% expect to use AI within six months.
The first two-week survey period starting in each month. Standard errors are about 1.2–2.5 points. Since late 2025 the survey question refers to AI as "super intelligence (formerly AI)".
View data as a table
| Nov '25 | 29.9% |
|---|---|
| Dec '25 | 30.5% |
| Jan '26 | 31.7% |
| Feb '26 | 29.6% |
| Mar '26 | 34.1% |
| Apr '26 | 33.9% |
| May '26 | 35.7% |
| Jun '26 | 34.3% |
| Jul '26 | 36.8% |
| Aug '26 | 36.5% |
| Sep '26 | 40.1% |
| Oct '26 | 39.2% |
Why advisors run out of time
Fidelity's 2025 Advisor Insights Study found that 59% of advisor time goes to administration, compliance and other non-client tasks. In an earlier Fidelity study, only 27% of advisors said they spend their time at work on high-value activities.
Hiring alone will not close the gap. The Bureau of Labor Statistics projects advisor employment to grow just 1% from 2025 to 2035, with about 17,100 openings a year mostly replacing people who leave, while Cerulli counts 105,887 advisors who plan to retire in the next decade.
- Meeting preparation pulls data from the CRM, custodians and portfolio accounting by hand.
- Notes, follow-up emails and CRM updates after every meeting take time no client pays for.
- Every outgoing message and marketing piece needs supervisory review.
- Onboarding means re-keying statements, IDs and transfer forms.
Deloitte projects that by 2032 AI could free 25–50% of adviser time from lower-value operational work. It is a model, not a measured result, but it shows where the opportunity sits: in the workflow around the advice, not in the advice itself.
AI use cases in wealth management
Most value today comes from removing the work around each client interaction. Pick an area to see where AI fits and which systems it needs.
Meeting prep briefs
Before each review, AI builds a one-page brief from the CRM (Salesforce Financial Services Cloud, Wealthbox or Redtail), custodian positions, portfolio accounting such as Orion or Addepar, and past notes.
Next-best action
AI spots money-in-motion events such as liquidity, inheritance or held-away assets and suggests which clients to call first.
Portfolio review narratives
AI explains drift against the investment policy and drafts the review commentary. The advisor approves every recommendation.
Tax and rebalancing insights
AI flags tax-loss harvesting opportunities and concentrated positions from lot-level cost basis data.
Client questions
An assistant answers routine questions about balances, transfers and distributions from custodian data, and hands anything else to a person.
Onboarding and document intake
AI extracts data from statements, IDs, trust documents and transfer forms and pre-fills account opening.
Service request triage
AI sorts the service inbox, answers status questions and routes requests to the right team.
Marketing review
AI checks marketing drafts against the firm's policies and the SEC Marketing Rule before a compliance officer reviews them.
Communications surveillance
AI reviews email and chat archived in tools such as Smarsh or Global Relay and surfaces the exceptions supervisors need to see.
Advisors are adopting AI fast, but not evenly
J.D. Power's 2026 study of 4,503 employee and independent advisors at US broker-dealers shows active AI use rising sharply in one year, with independent advisors still well behind.
- Employee advisors, 202673%
- Employee advisors, 202544%
- Independent advisors, 202642%
- Independent advisors, 202519%
Advisors who find their firm's AI tools effective report higher satisfaction: 781 points against a 632 average for employee advisors, and 826 against 688 for independents.
View data as a table
| Employee advisors, 2026 | 73% |
|---|---|
| Employee advisors, 2025 | 44% |
| Independent advisors, 2026 | 42% |
| Independent advisors, 2025 | 19% |
Not sure which workflow to automate first?
We map your CRM, custodian and compliance stack and show which AI use case saves advisor time fastest.
Where the data disagrees: do clients want AI or a person?
Clients say they expect AI, and many say they would trust it. The same surveys show that advisors and clients still see trust as a human job.
Clients are ready for AI
- 60% of wealthy clients expect their wealth manager to use AI, and 71% believe or suspect firms already do.
- 43% are open to AI delivering financial planning without a human advisor.
- 38% of advisors say clients already compare their advice with what they get online or from AI.
Trust is still human
- 68% of advisors say long-term client trust requires a human touch.
- Only 38% of clients trust AI as much as or more than a human advisor.
- 52% are open to AI listening to and augmenting their conversations with advisors, so almost half are not.
Evgeny Lupanov
Chief Technical Officer
Both are true, and they point the same way. Clients do not want a chatbot instead of their advisor. They want an advisor who answers faster, knows their file and has time to think. That is what AI should buy back: hours spent on notes, prep and paperwork, returned to the client relationship.
[ Regulation ]
What regulators have said about AI in wealth management
No AI-specific SEC rule
In June 2025 the SEC withdrew its proposed rule on predictive data analytics and said any future action would need a new proposal. In practice, firms apply their existing fiduciary, supervision and recordkeeping obligations to AI tools.
Emerging technology is an exam priority
The SEC's fiscal 2027 examination priorities list emerging technologies, including what the SEC now calls super intelligence, among the areas its examiners will focus on.
FINRA expects supervision of AI
FINRA's 2026 oversight report says summarization and information extraction is the top generative AI use at member firms, and flags agents acting without human validation as a key risk.
Our advice: treat AI notes as records
Get client consent, check state call-recording laws, keep transcripts and notes in the books-and-records archive and review them like any other client communication.
“In wealth management the model is the easy part. The work is connecting it to the CRM, the custodian feeds and the compliance archive, and keeping every number it shows traceable to the source. Start with meeting notes or client briefs, measure the hours saved per advisor, then move closer to advice.”
[ Case Studies ]
AI and investment platforms we have built
Akveo builds AI and data platforms for fintech and investment firms.
Wealth ManagementAI Development
Financial Multi-Agent AI Platform WealthFluent
Launched in 6 months with zero AI hallucinations
For a US fintech, we built an AI assistant that explains portfolios and goals in plain language. A separate deterministic calculation engine makes sure every figure comes from verified data, so the assistant stays compliant with US financial regulations.
InvestmentRetool
Investment Platform for Global Hybrid Capital Fund
Delivered in 2 months across 3 regions
For a fund investing in emerging-market fintech lenders, we integrated Retool components into its investor portal to centralize application pipelines, aggregate unstructured lender metrics and standardize risk evaluation.
Sources
- Fidelity Investments, 2025 Advisor Insights Study and 2024 Industry Trends Study · 479 advisors, February 2025; 432 advisors, February 2024
- Cerulli Associates, U.S. Advisor Metrics 2024 · Industry headcount model and advisor surveys, January 2025
- US Bureau of Labor Statistics, Occupational Outlook Handbook: Personal Financial Advisors · 2025–2035 projections
- J.D. Power, 2026 U.S. Financial Advisor Satisfaction Study · 4,503 advisors, December 2025–April 2026
- US Census Bureau, Business Trends and Outlook Survey · Finance and insurance sector, November 2025–October 2026
- Deloitte, Agentic AI could boost wealth adviser productivity · Deloitte model, May 2026
- EY, 2025 Global Wealth Research Report · About 3,600 wealth clients in over 30 geographies
- Edward Jones and Morning Consult, advisor AI survey · 201 financial advisors, May 2026
- Morgan Stanley, launch of AI @ Morgan Stanley Debrief · June 2024
- FINRA, 2026 Annual Regulatory Oversight Report · December 2025
- US SEC, Fiscal Year 2027 Examination Priorities · October 2026
- US SEC, withdrawal of the predictive data analytics proposal · June 2025
[ FAQ ]
Frequently asked questions
Discuss AI for your wealth management firm
Tell us about your advisors, your CRM and custodian setup and the workflow that takes the most time. We will come back with the AI use case that pays off fastest and a rough estimate.
Dmitry Klim
Head of Growth
