Do You Need to Tell Your Clients You’re Using Generative AI?

Whether you’ve already taken the plunge on implementing generative artificial intelligence into your practice or are still testing the waters, there’s one key question to ask yourself: What do I tell my clients about AI?

It’s natural to feel a little antsy about disclosing your AI use. Clients’ potential grievances are a valid concern, given a well-documented disclosure effect with AI, in which people tend to have an adverse initial reaction when it’s disclosed that something has been created by AI. For example, this disclosure effect has been found to make people more skeptical of AI-created news content and less willing to accept AI-made decisions. It can also mean they’re not willing to pay as much for an advisor who uses AI.

Though it may seem like a minefield, it’s still important to disclose your generative AI use to clients.

For one, our research finds that investors want transparency from advisors about their generative AI use. Fortunately, our preliminary data on the topic suggests that when clients are aware of their advisors’ generative AI, they tend to be more comfortable with it. Therefore, disclosure about generative AI may provide an avenue for building trust with clients—by demystifying what is being done with it, how it’s being done, and what it means for the client.

Additionally, consensus is emerging that disclosure is part of using AI ethically. And although there is little by way of legal guidelines currently, this is likely to change, and it behooves your practice to stay ahead of the curve.

Given the importance of disclosing generative AI, you may wonder how to do so in a way that can help mitigate the disclosure effect.

This principle pertains to several features of your disclosures. First, the content of your disclosure should be clear about what activities are AI-supported in simple, jargon-free language so that clients don’t have to guess what has been helped with AI and what hasn’t. Second, disclosures should be readily available to clients. That is, meet your clients where they are at. This may mean having shorter disclosures at the bottom of AI-supported content and linking to a more in-depth description of how you use AI. You may also consider providing clients with an FAQ page on the topic they can reference. In general, reducing the barriers to understanding your AI use may help clients be more receptive to the topic.

People can be hesitant to trust output from generative AI, so consider implementing a process for having a human check the work. Be sure to check and refine the AI’s output before it gets to a client and make clients aware of what this practice looks like (you can also include this in your disclosure). Another common concern people have about generative AI is the safety of their data and privacy. Using licensed versions of AI tools gives you contractual protections for the data you provide the tool, and including this information in your disclosure can help assuage client concerns on the issue.

Even with thoughtful disclosure statements and generative AI policies, you may encounter some clients who are just not comfortable with the technology. In these cases, rather than risk the possibility of alienating these clients, you should consider providing clients the opportunity to opt out of generative AI being used with their accounts. Don’t worry that this option will encourage too many clients to opt out: Most people tend to go with the default option. By having people choose to opt out (instead of in), only clients who truly care about not having AI used will say so.

Though tackling the issue of disclosing AI use to clients may be initially scary, doing so well gives you the opportunity to build trust with clients as you navigate the new AI-assisted financial planning landscape together.

The author or authors do not own shares in any securities mentioned in this article. Find out about Morningstar’s editorial policies.