TD Securities launches AI assistant for equity teams using OpenAI tech

TD Securities has launched the TD AI Virtual Assistant, powered by OpenAI’s GPT models and developed with Layer 6, a Canadian AI company acquired by TD Bank in 2018, to deliver real-time equity insights to its institutional sales, trading, and research teams. The assistant, which went live on July 8, represents a significant step forward for AI adoption in the highly regulated financial services sector, where banks have traditionally taken a cautious approach to implementing AI agents.

What you should know: The TD AI Virtual Assistant serves as a sophisticated knowledge management system designed specifically for front-office equity professionals.

• The assistant uses retrieval augmented generation (RAG) processes to access internal research documents, market data, 13F filings, and historical equity data.
• It integrates with TD Bank’s foundation model, TD AI Prism, which launched in June and processes 100 times more data than the bank’s previous single-architecture models.
• One standout feature is its text-to-SQL capability, which converts natural language prompts into database queries for easier data access.

How it works: TD Securities developed patent-pending optimizations to train the assistant without requiring fine-tuning of the underlying OpenAI model.

• “With patent-pending optimizations in prompt engineering and dynamic few-shot examples retrieval, we successfully achieved the business’s desired performance through context learning,” said Dan Bosman, TD Securities’ chief information officer.
• The system aggregates and synthesizes information into “concise context-aware summaries and insights” to help sales teams answer client questions.
• The assistant is designed to understand trading floor lingo and context, making interactions feel natural and intuitive for users.

Why this matters: The launch demonstrates how major financial institutions are moving beyond pilot programs to deploy AI agents in customer-facing operations despite regulatory constraints.

• TD Securities joins other banks like BNY, Wells Fargo, and Capital One in offering multi-agent solutions to sales teams.
• The assistant originated from an employee suggestion through TD Bank’s innovation platform TD Invent, highlighting how grassroots innovation can drive enterprise AI adoption.

What they’re saying: Bosman emphasized the collaborative nature of the project and its potential for external expansion.

• “What I’ve loved most about this is when you build something super magical, you don’t need to go out and sell or put a face on it. Folks come in and say to us, ‘we want this, we need this or we’ve got ideas,'” Bosman said.
• “My vision is that we see AI as something that can add value to us, but also to external customers at the bank. Right now, it’s a massive opportunity for us around driving a stronger client experience and delivering a better colleague experience,” he added.

The big picture: Financial institutions are gradually embracing AI agents after months of pilot testing, balancing innovation with strict data privacy and fiduciary responsibilities.

• The challenge lies not in teaching employees about AI tools like ChatGPT, but in establishing best practices, integrating agents into workflows, and understanding limitations.
• Banks face the additional burden of regulatory compliance while trying to deliver AI-powered solutions that can compete with more agile fintech companies.