AI Agents and the Future of Wealth Management

This newsletter is part of The Assetizer, GenTwo's thought leadership platform.


 

In September, Anthropic launched Claude for Financial Advisors, a suite of AI tools designed to support wealth advisors in their daily work. This followed an announcement from OpenAI, whose ChatGPT for Financial Services product provides similar types of tooling for investment banking and equity research. 

It comes of course as no surprise that agentic AI is coming to wealth management. The industry has long been ripe for automation, with high complexity and manual processes that make easy pickings for clever bots. 

At the beginning of the year I asked what radical automation might mean for wealth management as part of my six (hard) questions for 2026 post. With the agents now coming online, it seems like a good time to revisit the issue. 

What Has Landed 

To start, let’s look at what has been released. 
 
According to Anthropic, Claude for Financial Advisors is “a suite of connectors and workflow skills designed to help with research, prep, and documentation tasks.”  

On the connector side, the suite launched with an impressive list of partners, with names like BlackRock, Charles Schwab, iCapital, JP Morgan and Vanguard jumping out among a host of other well-known custody and portfolio data providers, asset management firms, alternatives platforms, market data and research houses, and workflow and client intelligence products. 

On the workflow side, there is a list of “skills” (instructions) for the agents to help advisors throughout their day, from client onboarding and compliance through meeting prep and to things like tax and estate planning. Such skillsets, as users of Claude or other LLMs will know, are simply plain-language instructions for the agents on how to carry out specific tasks. They can be run out-of-the-box or easily adapted to fit a business’s style of working. No programming knowledge is needed. 

And Anthropic’s new product is just one example. Besides the investment-banking oriented toolset released by OpenAI and mentioned above, this year has seen Raymond James roll out an in-house operations agent in January, Altruist add AI tax planning to its Hazel platform in February, Salesforce bundle six agentic capabilities into its financial services CRM in June, and Orion open its Denali AI to the thousands of advisors on its platform. That is the advisor side, where the agents work for the professionals. On the client side, where they work around them, Robinhood opened its platform to third-party trading agents in May and followed up in September with its own in-app agent that trades around the clock, Schwab promised AI-generated portfolio commentary for the millions of clients too small to merit a human relationship manager, and Morgan Stanley said it would open its workplace wealth platform to autonomous agents. Bots on both sides of the table, then, and nobody has been waiting for permission. 

The Automation Traps 

By combining access to data with a sophisticated understanding of the business and the role of the advisor, a suite like this is clearly on another level compared to the kinds of tools advisors have been used to working with. They can truly automate a sizable portion of an advisor’s daily work.  

Many in the business will naturally ask themselves where this all stops. Anthropic’s narrative is their tools are all about what I like to call “enhancement” – helping the advisor be more efficient so he or she can spend more time with clients.  

But what’s to say that such tools won’t lead to partial or full "replacement" of advisors and maybe even the advisory business as a whole? Here are some things to be worried about: 

• The efficiency trap. Efficiency can be a double-edged sword. It reduces costs but can also put pressure on revenue by inducing a race to the bottom as firms compete on fees. An enhanced, highly efficient advisor would also be able to handle a much larger client book than before. This will reduce the number of advisors that are needed, and likely thin the herd in favor either of the most talented ones or those who are first or better at using the agents. 

• The hidden money trap. Many advisory firms and brokerages earn significant revenue on things like idle cash, spreads on house products or retrocessions. Client-side bots will look to move idle money out of brokerage accounts into higher-yielding alternatives. The stock of Charles Schwab saw a selloff in September on exactly these concerns. 

• The trust in the machine trap. Over time, clients will increasingly trust the bots, putting more and more of their money into automated platforms. This will put increased pressure on the advisory business to justify itself (though perhaps not right away – see below). 

• The radical automation trap. Last but not least: The bots could potentially be used by clients to replace all money management, financial planning and investment advisory services. Wealth Management as a people business would slowly disappear. 

Light In The Darkness 

The threat that AI will eventually be good enough to completely automate the wealth management industry, making all advisors (not to mention everyone else in the value chain) obsolete is remote – but it is real. Yet the same can be said for almost any industry these days as we all wait to see where AI will take us. 

In the meantime, we at GenTwo think that things actually look promising, at least for those in the industry who know how to make the best use of the new tools. Here are some reasons why we think so. 

• Humans will be obliged to stay in the loop for a long time. FINRA in its 2026 Annual Regulatory Oversight Report made it clear that using an agent doesn't shift responsibility when it comes to financial advice. For automated processes they recommend a narrow scope and permissions, audit trails of actions, explicit human checkpoints before execution, system-access controls and agent-specific supervisory processes. FINMA leaders are saying similar things in Switzerland. With money and livelihoods involved, wealth management can be counted as one of those core industries that we as a society will not feel comfortable handing over completely to the machine. 

• Evidence suggests clients will prefer a human in the loop too, certainly for now. A bit paradoxically, it seems that investors feel the same. I say paradoxically because there is a lot of research and surveys indicating how many on the client side, particularly the young, are becoming less loyal to their advisory firms and are keen on trying out new online tools for their portfolio management. Yet, at least according to two recent studies in the US (Betterment, Vanguard), AI seems to be increasing demand for human advice, not the opposite. It seems people like using the machines to help them better understand their options. But they still rather trust a human when it comes to making an actual decision. 

• There will be great advantages for those advisors and firms who learn best how to enhance themselves. Those who can use agentic AI to bring down their costs, provide better services and take on more clients stand to reap great benefits. The chief of these will be time. People tend to be willing to pay a premium for trust and personalized service. Those advisors who put more time into their client relationships, and thus prove their value, will we think be better able to defend their fees.  

The Age of the Independents 

The wealth management industry has always evolved and changed. But most would agree that this is a time of particularly fast and deep change. We think right now represents a golden opportunity for independent advisors especially.  

Our Assetization thesis is that new technology and new approaches will level the playing field in the investment industry. Independent, boutique, and smaller operators will have the same capabilities as larger ones. AI agents are just part of this. The investment industry value chain is modularizing, standardizing and above all opening up. That puts a premium on brains and talent as opposed to muscle and might. For those endowed with these properties, the future should look bright. 

Tom Lyons, GenTwo