Financial Advice vs. Financial Influence in the Age of AI
By AG CPA Co. | Perspectives on Finance & Technology | ~650 words
There is a legal distinction, carefully maintained by decades of securities regulation, between providing financial advice and providing financial information. An investment adviser — registered, regulated, owing a fiduciary duty to clients — gives advice. A news article, a market commentary, a general educational resource gives information. The distinction matters because advice triggers obligations: to know the client, to act in the client's interest, to disclose conflicts, to be accountable for the recommendation.
AI systems are making this distinction increasingly difficult to maintain.
The Regulatory Framework
The SEC has been direct on the underlying principle: the fiduciary duty applies regardless of how advice is delivered — human or algorithmic. An AI system that takes a user's stated financial situation and generates a personalized investment recommendation is not providing general information. It is providing advice, and the entity deploying that system carries the adviser's obligations.
The American Bar Association's April 2025 analysis of robo-advisors and AI-driven investing made the corollary plain: AI-powered robo-advisors must meet the same standards as human advisors, including ensuring recommendations align with client interests, providing transparent and explainable disclosures, testing for algorithmic bias, and implementing robust cybersecurity measures. Robo-advisor assets are projected to reach $4.6 trillion by 2027. The scale at which these obligations now operate is significant.
The SEC added AI to its 2025 examination priorities. FINRA's 2026 Oversight Report introduced a dedicated section on generative AI, covering governance, recordkeeping, and autonomous agents. The White House's March 2026 National AI Policy Framework explicitly favored existing sector-specific regulators — the SEC and FINRA — over a new federal AI body, meaning that existing financial services law is the operative framework, applied to an entirely new class of tools.
The Influence Problem
What the regulatory framework has not fully resolved is the category of systems that sit between advice and information — systems that shape behavior without quite crossing into formal advisory territory. A mobile application that shows a user's spending habits against peer benchmarks is providing information. But if that information is selected, sequenced, and presented in ways designed to nudge the user toward specific financial products, it begins to function more like influence than information.
The SEC surfaced this concern in a 2023 proposal, which captured not only AI but also algorithms, machine learning models, and other analytical tools that drive personalization. The proposal signaled that technology-driven personalization could raise structural conflicts of interest that disclosure alone may not adequately address. The proposal did not ultimately advance in its original form, but the underlying concern remains active.
A 2021 German case illustrates the stakes. Regulators shut down an investment platform after discovering that its AI-driven tax optimization engine had provided unsuitable advice without proper disclosures or licensing. The absence of clear accountability left clients with limited recourse and required the return of millions of euros in assets.
The Practitioner's Responsibility
For financial professionals — CPAs, financial planners, investment advisers, tax attorneys — the practical implication of this landscape is that the tools they use, and the tools their clients use, carry regulatory meaning. A CPA whose engagement includes tax planning recommendations that incorporate an AI system's analysis is not absolved of advisory responsibility by virtue of having used software. Regulators evaluate the activity rather than the technology. If an AI tool contributes to personalized recommendations, it falls under the adviser's fiduciary obligations, supervisory controls, and recordkeeping requirements.
What this means in practice is that understanding how the tools work — not just using them — is part of professional competence. An adviser who cannot explain how a system generated a recommendation is in a difficult position if that recommendation is later questioned. Transparency is not only an ethical value in this context. It is an operational requirement.
The broader shift underway is not that AI is replacing professional judgment in financial services. It is that AI is changing what professional judgment needs to encompass. Understanding the technology, its limitations, its regulatory status, and its potential for influence — not just its analytical output — is increasingly part of what it means to serve clients well.
REFERENCES
SEC / Terms.law. 'AI Fiduciary Duty: SEC Robo-Adviser Compliance.' 2025.
American Bar Association. 'AI-Driven Investing: What Lawyers Should Know About Robo-Advisors.' April 2025.
Zocks. 'AI Compliance Guide for Financial Advisory Firms.' May 2026.
InnReg. 'SEC Guidance on AI: Rules, Alerts, and Enforcement Signals.' May 2026.
Arxiv / Robo-Advisors in Financial Services (ScienceDirect). January 2026.
Dialzara. 'Client Risk Profiling AI in Advisory Services: Robo-Advisor Challenges for 2025.' December 2025.
AG CPA Co. — Confidential