AI now permeates nearly every aspect of our personal and professional lives. Technology that was once considered “emerging” is transitioning to mainstream. Certainly, in financial services, work that a short time ago would have taken significant time and human capital to knit ideas together, can now be completed in an instant.

However, efficiency gains are not the holy grail of these tech advancements. And if the hype around generative AI proves even remotely true, there will be a levelling of the playing field, making it more challenging for firms to differentiate on technical capabilities alone.

How, in that case, can financial services institutions maintain a competitive edge? How will they find their differentiator of the future?

We believe it will be found through harnessing “collective intelligence”: the situational interaction of people and AI. Creating a collaborative and controlled environment for talent and tech to interact is essential to unlocking the transformative power of emerging technology.

Organizations who are reaping the benefits in this space are finding success at the intersection of hiring the right talent, investing in the right technologies, and blending the two to identify and address the right use cases. We believe the future isn’t artificial or human intelligence; it’s collective intelligence: where humans and AI work in partnership.

Think Differently About Talent

We know that talent is a critical ingredient in any successful innovation recipe. In our industry, deep subject matter expertise paired with advanced technical capabilities can be a major accelerant of transformative outcomes. These outcomes are especially important during major industry changes, as we are seeing now with the move to shorter settlement cycles in the U.S., Canada, and Mexico.

In our recent article “Recipe for Success in Transformation”¹ we outlined the importance of building cross functional, innovative teams to create a sustainable model for innovation. And, now there are supercharged team members in that recipe: Generative AI (GenAI) and Large Language Models (LLMs).

There are obvious benefits derived from marrying subject matter experts and technology. For example, LLMs that are trained by a company’s highest performers can lead to models that are infused by domain expertise and experience that can be leveraged by all employees to improve productivity, increase efficiencies, and accelerate training. A recent McKinsey report stated that “Generative AI has more impact on knowledge work… than on other types of work.”² Tasks that were previously thought of as too complex or nuanced for automation are now prime candidates for improvement.

How It’s Done Matters

In our experience, the selection of high value use cases is critical to optimizing investment decisions. And the key to the effective application of AI is to centralize its oversight/development and governance as a discipline and democratize its practical use across business teams. Putting AI directly into the hands of the business users, with a robust governance framework in place, allows solutions to be tested and enhanced by those closest to the operational challenges.

Democratizing AI across the organization leads to teams that think differently about their work, that think beyond the task they currently perform to the end-to-end workflow and how to incorporate AI/ML in both automating processes and augmenting decision making. They are no longer just clearing exceptions. Through interacting with the algorithms, they are labeling data and providing feedback to further improve the accuracy of the models, which leads to greater confidence in the predictions and further efficiencies.

As an example, at BBH we paired technologists and business teams to apply artificial intelligence to unstructured trade messages to reduce the need for manual intervention and processing. This work was foundational in order to create algorithms that predict which trades have a high likelihood of failure - these kinds of predictive analytics are now the table stake in a CSDR and T+1 world. So, this not only improves the oversight of the settlement process but also underpins new client analytics enabling them to manage counterparties and brokers more effectively.

LLMs may make it even easier to effectively connect those predictive trade analytics, in real-time, with other relevant processes (such as cash management, securities lending) to understand the potential impact, and actionable insights needed at the transaction level. Instead of combing through data and reports to link, now it may be possible through natural language interrogation. Now we’re moving from process optimization to process and product revolution.

With GenAI at the ‘peak of inflated expectations’³ on Gartner’s Hype Curve, the industry should continue to approach transformation through a pragmatic lens – selecting and proving out practical uses of the technology, following proven governance frameworks, and centralizing the tools through a Center of Excellence approach.

AI will not replace humans, but firms that can harness the collective intelligence of AI with humans will accelerate their service, offerings, products, and will have the market advantage and longevity.⁴