LTX, a subsidiary of global Fintech leader, Broadridge Financial Solutions, Inc. (NYSE:BR), today announced the launch of BondGPT+, the enterprise version of its BondGPT application, which incorporates Broadridge’s unique patent-pending technology and OpenAI GPT-4 to answer complex bond-related questions and rapidly assist users in their identification of corporate bonds on the LTX platform.

BondGPT+ contains a range of new capabilities focused on the needs of its most sophisticated enterprise clients.

“Since the successful launch of BondGPT in June 2023, we have received overwhelmingly positive feedback and thoughtful suggestions from our clients,” said Jim Kwiatkowski, CEO of LTX. “Based on that feedback, we have moved quickly to incorporate clients’ own data as well as a range of new features in the second generation of BondGPT. The result is that BondGPT+ not only meets our clients’ enterprise requirements, but also sets a new standard for generative AI solutions in the financial industry by providing a single natural language interface for seamless access to much of the data used by busy fixed income professionals.”

BondGPT+ allows clients to securely integrate their own data and other third-party datasets, providing easier access to critical information. Offering multiple implementation options and a robust API suite, BondGPT+ seamlessly integrates into clients’ enterprise applications, including trading workflows. This means businesses can harness the full spectrum of BondGPT+’s natural language capabilities to receive quick answers to complex questions, which can help clients optimize portfolio strategies and make better-informed trading decisions when time is of the essence.

The personalized BondGPT+ dashboard with resizable answer panels allows users to take up as much or as little screen real estate as required. Critical questions can be saved as “favorites” on the dashboard and scheduled to run at a desired frequency, time of day, or when key data changes. Users can tailor BondGPT+ to their requirements by adding preferences that should be considered when generating responses.

From the first instance, BondGPT was created with architectural design patterns that ensure accuracy, data security, and privacy. These principles are equally central to BondGPT+ enterprise implementations, and they are especially critical given the capabilities within BondGPT+ to integrate with proprietary client data and systems.

Because it was purpose-built for institutional fixed income users, accuracy is paramount in BondGPT, and BondGPT+ continues this heritage. The application relies on verified, curated data sources, which mitigates hallucination risk and eliminates the need to pass sensitive data through the LLM. Further, explainable AI capabilities allow users and administrators to understand how BondGPT+ arrives at its responses. The patent-pending technology behind the “Show your work” feature offers transparency by showing, step-by-step, how each response was created.

Another key differentiator of BondGPT+ is its focus on enterprise control and governance. With its Admin Center, system administrators gain the ability to configure various levels of data access through user and firm level entitlements. BondGPT+ builds on BondGPT’s AI-powered compliance layer, allowing compliance officers at client firms to add custom rules based on their firm’s unique compliance and risk management processes. AI-powered usage analysis provides invaluable insights into user interactions with BondGPT, offering businesses a powerful tool for improving user engagement.