A registered AI advisor is now free. Your moat is no longer the allocation.

What actually launched

In mid June 2026, Coinbase released a free AI investment adviser for its subscribers. The detail that matters is the word registered. This is not a chatbot that offers general commentary and a disclaimer telling you to see a professional. It is an SEC-registered investment adviser, which means it can give specific, personalized investment advice inside a regulated framework. Barron's, Crypto Briefing, and TechTimes all covered the launch around June 17, 2026.

For a working advisor, two facts in that sentence deserve weight. The first is free. A registered advice product at no cost, available to a mass market audience, sets a new floor for what people expect to pay for an answer to "what should I do with this money." The second is the line that ran through nearly every piece of coverage. The client still bears the risk. The AI can recommend. It does not stand behind the recommendation.

That gap, between a recommendation and someone who stands behind it, is the whole story for you. It is also the most useful frame you have for the next client conversation.

What this actually commoditizes

It helps to be honest about which part of the job is under pressure, because pretending nothing changed is how firms get hollowed out slowly. A free, registered AI advisor competes directly with the commodity layer of advice: gathering basic information, suggesting a reasonable allocation, flagging an obvious tax inefficiency, and rebalancing on a schedule. That layer used to be a respectable amount of what some advisors charged for. It is now something a subscriber can get for nothing.

If your value proposition, said out loud, is some version of "I will build you a diversified portfolio and keep it balanced," you have a problem that has been coming for years and just arrived. None of that is hard for software anymore, and now a regulated version of that software is free.

This is not a reason to panic. It is a reason to be precise. The allocation was never where the money was. It was where the conversation started. The free product takes the opening line of your pitch. It does not take the part of the job that actually keeps clients for twenty years.

What the AI cannot do, and where you earn the fee

Here is the part no free model touches. An AI advisor optimizes. A human advisor is accountable, and the two are not the same thing.

• Judgment under ambiguity. A client's real situation is rarely a clean optimization problem. A father wants to help one child more than the other and feels guilty about it. A business owner's "risk tolerance" on a questionnaire collapses the week a deal goes sideways. You read the human facts behind the financial ones. The model reads the questionnaire.
• Behavior in the bad moments. The single largest driver of poor returns is not bad allocation. It is the client who sells at the bottom and buys at the top. The job of talking a frightened person out of a destructive move is the most valuable thing you do, and it happens in a phone call the AI is not on.
• Accountability. When the AI is wrong, the client bears the risk, as every June 2026 article noted. When you are wrong, you answer for it, in person, to someone who can fire you. That accountability is not a bug in your model. It is the product.
• The relationship over time. You remember that the client's mother had the same diagnosis. You know they will say yes to the plan and then quietly not do it unless you follow up. That continuity is the moat, and it compounds.

The free product makes a strong case for all of this by contrast. It is the cleanest argument you have ever had for what a human advisor is actually for.

There is also a quiet point about scope. The AI advisor answers the question it was asked. It does not notice that the client framed the wrong question. A person comes in asking how to allocate a windfall, and the real issue is that they are about to make an emotional decision about a family business. A model optimizes the windfall. You catch that the windfall is a symptom and the family business is the case. That reframing, the move from the question asked to the question that mattered, is not something a free product is built to do, and it is often the most valuable thing that happens in a meeting.

Make the moat visible this quarter

Knowing your value is not the same as a client seeing it. The work this quarter is to make the human moat legible to people who can now get an answer for free somewhere else.

• Rewrite how you describe yourself. Cut every sentence that sounds like portfolio construction. Replace it with the moments you actually earn the fee: the talk-down in March 2020, the second opinion before a big irreversible decision, the plan that survived contact with a real family.
• Put accountability in writing. Tell clients, plainly, that a free tool gives them a recommendation and they carry the risk, while your advice comes with a person who is responsible to them. Say it before they read it in a headline.
• Show the difference between an answer and advice. The next time a client brings you something an AI told them, do not dismiss it. Take it seriously, then show them the three things it did not ask about: the spouse who disagrees, the tax lot from 2009, the goal they never wrote down. That demonstration is worth more than any brochure.
• Use the AI yourself, out loud. The advisors who look strongest here are the ones who use these tools to prepare faster and then add the judgment on top, in front of the client. AI for Financial Advisors is built around exactly that division of labor, where the model does the drafting and you keep the judgment and the relationship.

A useful companion to this is the workflow side. If you want the model to handle the prep that frees you up for the human work, better client recaps in six minutes walks through one concrete version of it.

The skill under the model

There will be another free advice product after this one, and another after that. The pattern does not change. Each launch commoditizes one more piece of the mechanical work and leaves the human work standing in sharper relief. The advisors who lose are the ones whose value was the mechanical part. The advisors who win are the ones who already moved their fee onto judgment, accountability, and the relationship, and made that move visible to the people paying them.

That is a method, not a reaction to one headline. If you want the structured version of it, AI for Financial Advisors teaches how to use these tools without ceding the part of the job that clients actually keep you for, and the two minute course quiz will point you to the right starting place. For the boundary on compliance and where the model has to stop, see where Claude stops and judgment begins.