Site icon Elevate Business Pro

Considerations For The Financial Industry

Considerations For The Financial Industry

Tamara Kostova, CEO of Velexa, empowers institutional clients through customized and embedded investing services.

Many companies were quick to append the acronym “AI” to any and every service remotely related to artificial intelligence after OpenAI’s public launch of ChatGPT at the end of 2022. Suddenly, these services were incorrectly being called “AI” in an effort to jump on the bandwagon.

The frenzy prompted the Federal Trade Commission to issue a stern warning in February 2023 about labeling products as “AI” when they weren’t. The FTC ended the warning with the tongue-in-cheek statement, “You don’t need a machine to predict what the FTC might do when those claims are unsupported.”

Being the CEO of an investing platform, I believe the concerns raised about generative AI—such as privacy concerns, copyright infringement and inherent biases—are quite legitimate. The financial sector has been using a form of AI since the 1980s to help with investing, financial planning and more. In 2008, the first consumer-facing robo-advisors were launched. But to confuse this mathematical technology with GenAI is misguided.

Robo-advisory isn’t the same as generative AI.

GenAI is a different application of AI technology than the mathematical models and machine learning algorithms that financial institutions have been using in robo-advisory. Both technologies can be called “AI,” just as an orange and a banana can both be called a “fruit.” But, internally, GenAI and the AI used for robo-advisory are different.

To get into the precise details of how GenAI differs from robo-advisory on a technical level would require a book, but from a high-level view, robo-advisor algorithms have been designed to follow investment best practices, and they typically follow a passive-indexing strategy. However, different robo-advisors will implement different proprietary algorithms.

GenAI like ChatGPT uses large language models. LLMs tend to struggle with mathematical computations, as they aren’t primarily designed to perform mathematical operations. Instead, GenAI is intended for generative functions, such as generating contextually relevant responses to user prompts. LLMs work through something called neural networks. These neural networks remain somewhat of a black box to designers. AI has also been found to hallucinate and produce erroneous outputs.

That said, robo-advisors also aren’t always right, just as financial advisors aren’t always right. From my observations, the errors of robo-advisory typically come down to incorrect algorithms and the inherent unpredictability of financial markets, despite our best efforts to attempt to predict them. Statistics predict probable outcomes, never guaranteed ones.

By lumping robo-advisory technology with GenAI, I believe we risk closing the door on underserved investors for whom robo-advisory opened the door in the first place.

Exactly how generative AI is used in robo-advising matters.

Still, GenAI has its use. Coupled with existing, mathematically precise robo-advisory AI, many powerful tools become possible.

For example, Morgan Stanley was the first major financial player to announce an OpenAI-integrated chatbot for use in financial advising. The bot was created for the company’s financial advisors and their support staff. It will provide advisors with “access to a database of about 100,000 research reports and documents,” CNBC said.

JPMorgan’s GenAI-powered robo-advisor is more ambitious. The financial services giant has registered a trademark called IndexGPT and plans to release its bot directly to consumers.

Given the financial sector’s heavy regulatory ecosystem, clearly demarcated guardrails and the reputation of these two companies, it’s hard to imagine that any GenAI implementation would be done in such a way as to bring them legal battles in the future. Presumably, both these tools will either have extremely clear warnings on them or they’ll be programmed to augment, rather than replace, existing, functioning AI technology.

From my perspective, the same is not necessarily true about startups that are hungry for market share, especially given the challenges robo-advisor startups have faced when trying to disrupt the industry. Some of those startups might consider GenAI the silver bullet they’ve been waiting for to give them an edge over incumbents. In this sense, yes, regulation might be welcome, but where does the line for regulation vs. innovation in AI begin and end?

Robo-advisors helped open the door to an underserved sector.

Robo-advisors opened the door to an underserved sector of investors who did have some capital to invest but not enough to purchase professional advisory services. Collectively, that capital is immense and a much-needed source of revenue for larger funds seeking to make their revenue targets. Catering to this demographic individually would be impossible. I believe intelligently developed and responsibly implemented robo-advisory software can keep this demographic active in investing.

The other door opener is intelligent regulations. Grouping all robo-advisory services into a lump, whether they use GenAI or not, would be the biggest mistake. As I see it, doing so could stifle an industry that has been running successfully without using GenAI. Some professionals in the tech industry have said the European Union’s AI Act, for example, is too broad and “may catch forms of AI that are harmless.”

That’s a challenge. AI companies need to ensure regulators understand these nuances. Leaders can do so through joining trade bodies, putting out educational materials and inviting regulators to be part of roundtables or panels.

It’s also essential that those of us in the financial sector are critical when assessing AI-enabled partners. In addition to their feature set, it is important to investigate the fundamental technology and the types of strategies implemented for end users. This will also tie into the longer development of AI and what future strategies it could potentially implement. Another consideration is data privacy and security, as there is much sensitive financial information to be ingested.

AI is not just a black box we as leaders should simply accept for use; instead, it should be scrutinized and selected carefully, just like any other technology provider.


Forbes Business Council is the foremost growth and networking organization for business owners and leaders. Do I qualify?


link

Exit mobile version