Recent and rapid developments in generative AI are changing the world as we know it. Using Large language Models such as viral sensation ChatGPT and other language models based on generative AI, it is possible now to make entire songs, videos, marketing material, and yes even write articles – all tasks once exclusively performed by humans.
In one remarkable instance, AI's transformative power attained new heights as it ‘resurrected’ the voice of John Lennon, bringing the long dead musician ‘back to life.’
It is inevitable that generative AI will reshape the financial planning landscape. Now is the opportune moment to evaluate the impact of AI tools on the financial planning industry and its stakeholders.
ChatGPT is a Better Portfolio Manager Than Many
By all indications, artificial intelligence programs such as ChatGPT and Google’s Bard are already making big waves in the financial services market.
Industry heavyweights such as JP Morgan, Morgan Stanleyand Goldman Sacks, are already finding ways to incorporate the disruptive new technology in their workflow.
While some believe it may be too early to go all-in, there is at least one study that suggests ChatGPT's selection of assets, and portfolios constructed therefrom, is significantly better than randomly selected assets.
There Are Benefits to ChatGPT for Financial Advice
Generative AI has the potential to revolutionize the financial advice landscape, says Mehrnaz Shokrollahi, AI team lead at PureFacts Financial Solutions.
It has the potential for “offering advancements such as automated portfolio management, personalized financial offerings, and anomaly detections,” she says, noting that the program leverages algorithms to generate optimized portfolios based on an investor's risk tolerance and objectives, and offer personalized financial offerings to match individual needs.
Additionally, generative AI aids in detecting anomalies in financial systems, enhancing security and fraud prevention, she notes.
Is Your Financial Planner ChatGPT? Maybe a Little.
Some financial firms are dipping their toes by integrating generative AI tools into their investment advisory services.
“These AI systems process large volumes of financial data, market trends, and individual investor preferences to generate personalized investment recommendations,” says Shokrollahi. “This enables financial advisors to offer more tailored and data-driven advice to their clients.”
Other firms are using new AI tools to enhance their customer service channels by deploying virtual assistants and chatbots to handle inquiries and provide basic support including “account information, transaction details, and simple financial advice,” says Shokrollahi.
By automating routine tasks, financial institutions can improve response times, enhance customer experiences, and achieve cost savings.
Early adopters of generative AI are convinced it has the potential to enhance the capabilities of financial advisors and investors by offering value-added support and brining efficiency to tasks such as “market analysis and trend identification, investment research and portfolio management, anomaly detection, and compliance and regulatory adherence,” Shokrollahi notes.
All This Innovation Comes with Risks
As with any new technology, generative AI has its shortcomings and needs to be further fine tuned for better accuracy and improved efficiency. Therefore, it is important to understand the risk associated with these tools.
Shokrollahi particularly points out “issues related to data privacy and security, ethical use of AI, biases in AI algorithms, and the need for human oversight and accountability in critical decision-making processes.”
Another challenge inherent in AI is algorithmic bias, inconsistencies and inaccuracies. “If not properly analyzed, AI algorithms can inherit biases, potentially leading to unfair outcomes,” Shokrollahi cautions.
Another issue is accountability for advice. Who is ultimately responsible for the advice that is given? “I assume that will remain the financial advisor’s responsibility, just as it’s the advisor’s responsibility today for overseeing the computer-generated information that he or she provides,” assertsJohn Rekenthaler, director of research for Morningstar Research Services.
Perhaps the firms that sell AI services should also be held responsible, he notes.
ChatGPT is a Complementary Tool, Not a Replacement
Given these risks, financial service companies may limit the use of generative AI for general business purposes, such as customer service and call centres, says Morningstar director of equity research Michael Wong.
“There have been some stories of companies using generative AI for research,” he says, alluding to Morgan Stanley working on a generative AI program that their “financial advisors could use to summarize research that had been produced by” the bank’s analysts.
Financial experts see generative AI tools having a supportive role in the work of financial advisors for the foreseeable future. “AI could permit financial advisors to be far more efficient,” says Rekenthaler.
“AI would in effect be a personal assistant that does much of the work that advisors now do, leaving them with the responsibility of editing the AI’s suggestions and providing the final sign-off.”
By freeing up time spent on doing more mundane tasks, AI should permit financial advisors to serve more clients. “In the future, advisors are likely to be less rewarded for having in-depth knowledge of their subject matter, as AI programs can substitute for that knowledge, and are likely to be more rewarded for understanding how to most efficiently harness that technology in their practices,” argues Rekenthaler.
That said, Rekenthaler doesn’t rule out the possibility of a more sophisticated future AI model opening doors to an evolved AI role in financial planning. “Further down the line, AI will become intertwined with the financial planning process,” he says. “The advisor will retain the personal relationship, but AI will assist in asking the questions and will ultimately create the financial plans.”
The advisor, he adds, will bear the final responsibility for the plan that they recommend to the client. He, however, maintains “there’s no reason why a human would create plans that can be created more accurately, in the sense that the same inputs will aways lead to the same answers, by an AI routine.”
