SoFi AI Coach: Half of User Chats Are About Investing

Visual explanations can prompt behavior change where text falls short, Walsh said. A written statement that 30% of discretionary spending goes toward one category may have little impact. A pie chart showing that category towering over every other expense can create a different response. “There are a lot of financial concepts that are much easier to express visually than they are in writing,” he said. “It opens up this whole new dynamic of how do we communicate these topics in as simple and understandable manner as possible.”

The expansion of AI guidance raises a question: Where should automation stop? Coach was built using principles, communication methods and boundaries drawn from eight years and tens of thousands of conversations conducted by human financial planners, Walsh said. Rule sets are designed to identify when an interaction moves beyond the intended scope and should be escalated. “There are going to be clear rule sets that are built in,” he said. “This is going to expand beyond the scope of the advice or the guidance that we want to provide. Now let’s escalate this to a human interaction.”

That human-in-the-loop structure may be essential to scaling AI in regulated financial environments. It allows software to handle education, organization and routine guidance while preserving a pathway for complexity, judgment and regulated advice. The result could be a two-sided productivity gain. AI tools can serve consumers who might otherwise receive no guidance, while human planners can use the information gathered through digital interactions to serve more people efficiently. Walsh described the dynamic as a “snowball effect” where expertise reaches more people and the human side becomes more efficient.

For SoFi, the Coach is more than a feature. Deeper engagement through the platform could reduce churn and increase lifetime value, especially as the company competes with other fintechs and traditional banks for users. SoFi’s advantage lies in its integrated suite of banking, lending and investing products, which gives the AI a fuller picture of a user’s finances than a standalone robo-adviser might have. The escalation framework also addresses the liability question that hangs over automated financial advice, a topic explored in AI Rogue Agents: Who Bears Liability for Breaches.

Walsh said the ultimate measure of progress is whether technology can make informed, contextual and actionable guidance available to people who have historically been left outside the financial advice system. SoFi’s Coach is one attempt at that. Half the interactions already focus on investing, the hardest financial behavior to change.