Can financial advisors use Claude with client information?

Yes, with care. Turn off the model improvement setting in Claude's privacy settings, remove names and account numbers before you upload documents, and check your firm's compliance rules on review and archiving. For large volumes of regulated files, look at a Team or Enterprise plan.

Will AI send emails to my clients on its own?

Not in the way we set it up. The AI employee writes the email and saves it as a draft in Gmail. You read it, edit it if you need to and send it yourself, so your normal review step stays in place.

Does Claude work with my CRM?

Claude connects to a long list of tools through connectors, and many CRMs either have one or can be added as a custom connector. Luke runs GoHighLevel that way. If you don't use a CRM, a Google Sheet or an Excel file works fine to start.

What's the first AI employee an advisor should build?

The meeting follow-up email. It happens after every meeting, it follows the same pattern each time and it's easy to judge. Once it sounds like you, add the prep brief before meetings and connect both to your calendar.

Do I need to be technical to set this up?

No. Everything here is built in Claude by typing what you want in plain English. The AI Operating System is six live sessions that install these AI employees inside your business with you, so you leave with them running.

AI for financial advisors: what to set up first

AI for financial advisors

AI for financial advisors works best on the repeatable work around client meetings: prep briefs, follow-up emails, CRM notes and weekly reporting. In Claude you set each job up as an AI employee, trained on your process and connected to your email, calendar and CRM. Everything a client will see waits as a draft for you to approve.

Where AI employees help first

A prep brief before every review meeting

Before a client review, it reads the calendar invite, your notes from the last meeting and the client's record in your CRM, then hands you one page: what changed, what you promised to follow up on and what to ask. You walk in prepared without opening four tabs.

Claude Project
Google Calendar connector
Your CRM

The follow-up email after the meeting

You drop in your meeting notes or a transcript and it writes the recap email in your voice, with the action items and next dates spelled out. It lands in Gmail as a draft. Nothing goes to a client until you read it and hit send.

Claude skill
Voice skill
Gmail connector

A warm prospect list from people who already know you

In the Oct 13 masterclass Luke cross-referenced 8,000 Instagram followers against 2,000 contacts in his CRM to find the people who already follow him and have shared an email. An AI employee then drafted a personal email to each one. For an advisor, the same move finds the clients, referrals and event attendees who are already paying attention.

Claude
CRM connector
Gmail connector

A Monday summary of your book

A scheduled task runs every Monday morning, reads your calendar and CRM, and lists the week's meetings, the follow-ups still open and the clients due for a review. It's the list you'd otherwise build by hand on Sunday night.

Scheduled task
Google Calendar connector
Your CRM

What do financial advisors want AI to do?

We ask everyone who joins the Tuesday masterclass what they'd hand to an AI employee. The people in financial services, investment management and insurance who answered want help with the same few jobs: getting leads, prospecting, social media, admin and answering client questions.

Across the whole room it looks similar. Out of 174 business owners polled across 19 industries, the top requests were email management and replies, then social content, then reporting. For an advisor most of that work sits around the meeting, before it and after it, which is why the use cases on this page start there.

How do advisors keep client data safe with AI?

This is the question that comes up most when advisors and insurance agents are in the room. Here's the checklist Luke walks through:

In Claude's privacy settings, turn off the setting that lets your chats be used to improve the models.
Strip names, account numbers and other identifying details before you upload a statement or a plan.
Never paste passwords or logins into a chat.
If you handle a high volume of regulated files, look at a Team or Enterprise plan before you scale up.

Then check with your compliance team or broker-dealer on how AI-drafted emails get reviewed and archived. Because every email in these setups is a draft until you send it, it fits the review step you already have. Luke's own rule is to read the vendor's trust center before connecting anything, and Anthropic's is the one he trusts with his own data.

Where should an advisor start?

Start with the job you do after every single meeting. The follow-up email is a good first AI employee because you do it the same way every time and you can tell right away if the draft is good.

Then climb the four levels. Level 1 is a contractor that does a job with no context. Level 2 is trained on your process and your voice. Level 3 is connected to Gmail, your calendar and your CRM. Level 4 runs on a schedule, like the Monday summary above.

One advisor in the room put it this way after a session: "You just gave me time back. I have been trying to figure all of this out on my own." James Brewer, a financial advisor in the AI Operating System, said he'd already gotten his money's worth by session 3.