There’s no denying that artificial intelligence is developing quickly—at warp speed, even. In fact, in March 2023, some of the biggest names in technology—including Elon Musk and other professors, researchers, and business leaders—signed a letter asking for a pause for artificial intelligence labs training AI systems out of concern for the dangers such technology may present. Additionally, the United States and the United Kingdom have held high-level summits about AI safety in 2023.
Even with such concerns about what a proliferation of AI could mean for society as a whole when it comes to AI and wealth management, there’s a place for using tools based on technology. “When you’re looking for a statistical or high-level outcome or solution, it’s helpful,” says Whittier Trust SVP and Senior Portfolio Manager, Teague Sanders, who notes that quantum computers, such as the one built by Google, are approximately 158 million times more powerful than the supercomputers used today. That means that answers—from researching companies that may present investment opportunities to pulling numbers to analyze industry trends—can be at our fingertips more quickly than ever. Savvy client services advisors can leverage such technologies to inform expedient answers, recommendations, and reporting.
Fact-checking: a vital component for the use of AI in wealth management
Large language models (LLMs) are deep databases pre-trained on mind-boggling amounts of information. That’s why ChatGPT Bard, LLaMA, PaLM2, and many more have become popular tools for asking a question and waiting for an (almost instant) answer. While Sanders says that Whittier Trust has subscriptions to some LLMs because they can be useful for summarizing things and finding links and patterns, Whittier Trust team members always thoroughly double-check the results to verify the veracity of the information. Case in point: “There can be ‘hallucinations’ within a dataset,” Sanders explains. “If you ask an AI-driven LLM such as Bard a question like, ‘What was the revenue generated by X business?” and it gives you an answer you don’t think is right, it will re-generate a different response. You have to make sure that, when an LLM does a calculation, it’s analyzing the right thing and using authentic data.” Again, it comes down to focused human oversight.
AI and LLMs can also be useful for shaking up the thinking on a particular topic, sparking creativity and brainstorming. “If we have a client situation or an investment we’re considering, we might throw six or seven different prompts about a topic into one of the LLMs and just see what comes out,” Sanders explains. “That can be a catalyst for creative thought.” For example, if the team is thinking about an investment, an AI-driven tool can help. “If we’re looking at Mr. Carwash, we might ask the LLM to show us the entire landscape of car washes in the United States or the last six quarters of earnings of car washes in the American Southwest.” Such queries can help frame issues the team is considering, but it won’t be the deciding factor.
Why wealth management AI won’t replace a human touch
Artificial intelligence (AI) is everywhere, it seems, from predictive text in our Google searches and chatbots in customer service to facial recognition when we check in for a flight at the airport. Even though those things and more have become commonplace, there are some areas of our lives where such technology isn’t compatible: namely, the expansive use of AI in wealth management. “Wealth management, specifically, our style of wealth management simply doesn't lend itself to leaning heavily on AI,” says Sanders. “Even with all the advantages and efficiencies AI can bring, when you look at our clients and what we do for them, it’s really about the integration of our complete service offering; specifically our five pillars of expertise [family office, investments, philanthropy, real estate, trust services]. The comprehensive service these areas of expertise bring allows us to provide the personalized and compassionate approach that makes us successful, unique, and powerful for our clients.” The use of AI in wealth management can be valuable in some instances, while simultaneously complementing human expertise and intelligence for even greater results.
One of the primary ways Whittier Trust serves clients and sets itself apart from other firms is its highly personalized approach to serving the whole person or family. “An AI-driven solution does not exhibit empathy right now, it does not get to know someone’s hopes for family continuity or heart-felt goals for making a difference in a philanthropic endeavor. A computer doesn’t hold someone’s hand as they’re going through a difficult season,” Sanders says. “Those things require a lot of human touch because there's still a lot of emotional involvement when you're trying to come up with a customized, tailored solution for each very complex family.”