How RBC Borealis and RBC are creating value for customers through AI and Machine Learning.
RBC is committed to understanding and harnessing the power of AI for its customers and stakeholders. At RBC’s recent Financial Institutions Conference, we heard from three divisions within the bank, each of which is actively exploring how the application of AI and Machine Learning can continue to improve banking for the better.
At RBC Capital Markets, for example, the Quant Sales Trading & Strategic Initiatives team is actively looking at how AI can improve electronic trading with a time horizon of between two to five years. RBC’s Technology & Operations team is exploring ‘horizon 2’ innovations like generative AI and quantum computing. And at RBC Borealis, RBC’s AI research and development centre, about 120 researchers, engineers and product professionals work across the bank to see how AI can solve some of the organization’s biggest challenges.
Client First 👤
In part, the success of RBC’s various AI initiatives and activities comes down to a heavy focus on customer-centricity and collaboration across the bank.
“We need to remain 100% focused on the business problem we are trying to solve,” added Alex LaPlante. “As technologists, it’s easy to go off and build a really cool technology. But if it doesn’t solve an immediate and important business problem in the way our clients and partners need, then it’s not delivering value.”
It's not just at the research and development phase. RBC’s technology teams work with their clients through the lifecycle to drive adoption and continuous improvements. “We spend a lot of our time helping our traders really understand how our technologies can give them a better outcome and how it adds value to their trading process,” said Susom Ghosh, Global Head of Quantitative Sales Trading & Strategic Initiatives at RBC Capital Markets.
RBC Capital Market traders and RBC Borealis scientists continue to collaborate to deliver a real-world AI solution through the Aiden® platform to help improve trading results and insights for clients in a measurable and explainable way. The recent launch of Aiden® Arrival signals an important milestone in the expansion of our innovative Aiden® trading platform.
The business of trust 🤝
Another key to RBC’s success with AI is the organization’s commitment to develop AI responsibly. “We believe in being responsible by design where we get a wide range of experts around the table to assess our ideas and to help us identify the risks we should be thinking about right from the start,” noted Alex LaPlante. “That really defines how we move forward and the technologies and approaches we choose so our AI solutions respect diversity and human integrity.”
“It’s a key question for generative AI and large language models right now,” added Sal Vella, Vice President T&O Strategy, Innovation and Architecture. “We know these technologies could add value. But the question is, when do we deploy it? Maintaining the trust of our clients is paramount, so we really need to understand the risks involved around fairness, bias, transparency and privacy as we think about deployment.”
There are definitely some interesting questions being asked. For example, what does 95% accuracy feel like for a chatbot? Is that good or is that bad?” asked Alex LaPlante. “It really depends on where it fails – if it does a really great job answering your question but swears at you the whole time, that’s probably not the type of product you want to bring to market.”
Delivering the right solution ♻️
AI is not a silver bullet for all the business’s problems. “Whenever we kick off a new AI project, I ask myself two questions. The first is can we do it? The technology is maturing rapidly, but there are many problems that are still way beyond where we are today,” added Alex LaPlante. “The second question is should we do it? The reality is that AI is expensive to deploy and maintain, so you really need to know that the tradeoffs make sense.”
To view a replay of the panel discussion, click #RBCFI livestream on possibilities for AI-driven solutions or visit the RBC Financial Institutions Conference website to learn more.
News
RESPECT AI: Improving the efficiency of Differential Privacy with Zhiqi Bu, Amazon AWS AI
Machine Learning Reading Group: The Mechanics of n-Player Differentiable Games
News