TD Traders Leverage ChatGPT For Faster Market Insights
In the fast-paced world of financial markets, speed is a critical advantage. TD Securities is embracing this by deploying an AI-powered virtual assistant, transforming how its traders generate ideas and access research. Launched in June, this tool is already demonstrating its value by shortening research time from hours to mere minutes.
From Hours to Minutes AI Supercharges TD Traders
"It's a massive time save," Dan Bosman, Chief Information Officer at TD Securities, explained. He highlighted the daily challenge for capital markets professionals who often receive numerous lengthy research documents at once. "Even if you're the fastest reader and you're really skim reading, you have to take that first 30 minutes in the day to pore over that before you can make your first call. Now with the tool, you're able to get those insights and make those calls within minutes."
While some will still read the full reports later, the immediate need is to cut through the noise. The AI assistant helps traders quickly identify the most crucial signals from a constant barrage of content, allowing them to inform clients faster and more effectively.
A Bank Wide Bet on Artificial Intelligence
This initiative is a key part of a much larger AI strategy across the entire TD Bank Group. CEO Raymond Chun has set ambitious goals, aiming for $1 billion in annual value from AI through a mix of revenue growth and cost savings. "Across TD, we're deploying the capabilities needed to drive speed, such as AI-powered virtual assistants, AI-enabled adjudication, predictive tools and new applications," Chun stated.
The results are already tangible, with the bank approving mortgages in hours instead of days and pre-approving credit cards for millions of clients. With a team of 2,500 data scientists, engineers, and experts, TD is making AI fundamental to its business and client experience. This commitment extends to other areas, such as providing coding assistants to software developers within Bosman's own team.
The AI Arms Race on Wall Street
TD is a strong contender in the financial industry's push for AI adoption, but it is not alone. "Getting AI and generative AI right is a key initiative across broker dealers and investment banks across the Street," said Brad Bailey, research director at Burton-Taylor Consulting. He noted that TD's early acquisition of AI firm Layer 6 has helped it create value, but major competitors are also investing heavily in similar technologies.
The common goal for many banks is to unlock the vast knowledge buried within internal systems. Sumeet Chabria, CEO of ThoughtLinks, observed, "Generative AI tools like ChatGPT are proving very effective at surfacing this information in a natural, conversational way. Creating an access layer that makes research and data more discoverable and usable is becoming a common goal for banks."
Under the Hood How the Trading Bot Works
TD Securities selected OpenAI's enterprise version of GPT, leveraging its existing cloud partnership with Microsoft Azure. Bosman emphasized the importance of the platform's enterprise-grade security and privacy features. The virtual assistant was co-developed by Bosman's team, the sales and trading team, and Layer 6, the AI company TD acquired in 2018.
A key focus for the development team was training the AI to speak the language of finance. "It sort of sounds like an equity researcher and it sounds like a salesperson," Bosman noted. The tool uses a technique called retrieval augmented generation (RAG) to ensure its responses are based solely on the bank's own proprietary research and market data, preventing the AI from "hallucinating" or generating fabricated information.
Beyond answering questions, the bot can translate text into structured query language, allowing it to produce tables, graphs, and other data visualizations on demand. "Before, you had to have a business intelligence tool, and a data analyst to do the work," Bosman said. "Now, you can literally just ask for what you want to see, and you get it."
Real World Applications From Idea Generation to Macro Analysis
The practical use cases for the virtual assistant are numerous. Traders use it for meeting preparation, asking for quick summaries or even specific trade ideas with prompts like, "Give me five trade ideas." Many start their day by asking the bot for the five most important things they need to know.
The tool also excels at identifying broad, macro-industry themes that would be difficult for a human to piece together manually. For instance, a trader can ask about the sector-wide impact of a government shutdown or new tariffs, helping them uncover correlations and insights that provide a competitive edge for their clients.
Whats Next Agentic AI and Full Scale Deployment
Currently, around 1,000 employees at TD Securities are using the virtual assistant, with plans to deploy it across the entire global markets business soon. The project has been highly successful and has attracted interest from other divisions within the bank.
Looking ahead, Bosman is optimistic about the future, envisioning 2026 as "the year of agentic" AI. Agentic AI refers to systems that can autonomously execute complex, multi-step tasks. "I think there's immense value to be unlocked with operations and tech looking at how we can reimagine processes using this technology," he concluded.
