What Mattered This Week

The advisor tech stack is being restructured from the inside out. This week made that unmistakable. Envestnet unveiled its most comprehensive AI strategy to date at Elevate 2026, committing to AI-native workflows, explainability, and embedded decisioning across a platform that touches $7 trillion in assets. Orion's CEO shared a productivity thesis - doubling advisor capacity, potentially doubling the share of Americans who receive financial advice - and backed it with live product. A new Ezra Group report documented what advisors can already feel: the AI notetaker choice at firms right now is actually a choice about who orchestrates the entire advisor stack.

The deeper pattern is the race to own what strategists are calling the "system of action" layer - the AI that helps advisors decide what advisors do next, not just record what they did. Every major platform announced this cycle is competing for that position: Envestnet, Orion, Wealthbox, Jump, Zocks. No single platform dominates - yet.

Meanwhile, FINRA's first-ever GenAI governance section is the active compliance framework advisors are operating under right now, and AI-powered fraud has reached industrial scale. The same technology enabling advisor productivity is being weaponized against advisory firms.

Here's what moved this week, what it means operationally, and what to do about it.

AI TOOL ROUNDUP
The war for the advisor desktop is fully underway


  • At its annual conference, Envestnet committed to an AI strategy built on three pillars: orchestrated workflows, personalized next-best-action advice, and enterprise governance.

  • Envestnet introduced AI explainability features that translate dashboard data into plain-language insights, plus a deeper MoneyGuide integration that auto-identifies accounts and holdings from uploaded client files and feeds them directly into proposal workflows.

  • An AI-native platform commitment at this scale - $7 trillion in assets, thousands of enterprise clients - reshapes the baseline expectation for what advisor technology must now deliver.

  • Orion's Denali AI is now in production across its $5.8T platform, with two new assistants: Report Assistant for automated client reports and Query Studio for plain-English data queries - replacing the "swivel-chair" workflow across research, custodial, planning, trading, and reporting systems.

  • CEO Natalie Wolfsen named Orion’s big ambition: double financial advisor productivity so that the share of Americans receiving financial advice grows from 20% to 40%.

  • Advisors using Orion reported a 2025 organic growth rate approximately 40% higher than non-Orion peers (per Orion's own survey data); 17 of the top 20 Barron's RIA firms are Orion clients.

  • A new Ezra Group buyer's guide identified a fundamental bifurcation in the AI notetaker category: "Legacy path" tools feed data to the CRM; "Agentic OS path" tools orchestrate the entire advisor stack from above.

  • Jump and Zocks are firmly on the second path - Jump's AI Associate now executes actions across CRM, email, and planning tools from a single interface. Meanwhile Zocks is explicitly expanding from admin automation into surfacing revenue opportunities across an advisor's full book.

  • Evaluating a meeting AI tool is increasingly choosing what will become an orchestration platform, raising the stakes to a firm-level architecture decision, not just a feature purchase.

RISK REALITY CHECK
Governance is now a live examination priority

  • FINRA's 2026 Annual Regulatory Oversight Report introduced a dedicated GenAI section for the first time, establishing the examination framework that broker-dealers and investment advisors are being measured against.

  • The core requirements: test all GenAI tools for accuracy, privacy, and bias; log prompts and outputs with model version tracking; implement human-in-the-loop review; and conduct thorough initial and ongoing diligence on every third-party AI vendor - including what data they access and what AI they use internally.

  • FINRA explicitly named agentic AI as an "emerging trend" requiring specific governance attention, and reminded firms that all AI-generated client communications must meet the same fair, balanced, and not-misleading standard as human-written ones.

ADVISORS APPLYING AI
What deep AI adoption looks like

  • Mariner Wealth Advisors ($25B+ AUM) has deployed a fully layered AI stack on top of Orion: Zocks for meeting intelligence, Sherpas for AI investment proposals, and Humanity Labs for back-office automation.

  • 125 advisors are using Sherpas to take prospect conversations to personalized proposals in hours rather than days, per Mariner CIO Kenny Pointer.

  • The governance model is deliberate: advisors guide the strategy and sign off on the final output; the AI produces the analysis and structure. Human judgment stays in the decision.

What This Means

  • The system-of-action war has no clear winner yet. Envestnet, Orion, Wealthbox, Jump, and Zocks are all competing to own the layer that decides what advisors do next. Firms that understand this architectural question now will avoid a costly re-evaluation in 18 months.

  • Supply has lapped demand. Governance-ready, enterprise-grade tools are widely available. 23% of advisors still use none. The bottleneck is adoption. Advisors who move in the next 12 months have a real first-mover window.

  • Governance is a selection criterion, not an afterthought. Output logging, human review workflows, and documented vendor diligence are now FINRA expectations. Ask every AI vendor: how do you log outputs, what is your data retention policy, and who owns your compliance controls?

  • The next frontier raises the stakes. The current wave covered meeting notes and CRM updates. The new wave - Altruist Hazel tax planning, Sherpas proposals, Envestnet personalized advice - is AI closer to the advice itself. Hallucination and supervision risks are materially higher in that territory; governance architecture will determine which vendors earn trust at scale.

Until next week,
AI WealthTech Weekly

Keep Reading