Bloomberg: Using Gen AI for Enhanced Financial Data Analysis

The fintech industry has been steadily embracing generative AI technologies to process the immense volumes of data that financial professionals analyse daily.

While several major financial institutions have incorporated Gen AI assistants into their research processes, these early implementations often faced challenges with accuracy and interpreting the specialised language of financial documents.

Additionally, regulatory concerns regarding the reliability of AI-generated insights have limited adoption in certain market segments.

Addressing these issues, Bloomberg has launched AI-Powered Document Insights, an innovative tool leveraging Gen AI to help research analysts and corporate executives extract information from company documents through natural language queries.

This system represents one of the more ambitious applications of conversational AI to financial document analysis from an established market data provider.

Bloomberg's New AI Solution: Document Insights

The tool enables users to search and summarise financial documents using conversational queries, building on Bloomberg's 15-year experience applying AI to financial information.

Bloomberg Intelligence analysts contributed to training the Gen AI models to understand financial terminology and Bloomberg's AI guardrail systems are also in place to ensure accuracy and prevent hallucinations or fabricated responses.

This research solution allows clients to adjust investment positions by quickly analysing both text and structured data within company documents from Bloomberg's extensive content repository.

The library includes over 200 million company documents, more than 5,000 daily Bloomberg News stories on markets and economies, and research content such as Bloomberg Intelligence reports covering thousands of companies.

Industry Feedback

The tool has received positive feedback from finance professionals who value its time-saving capabilities.

Magdalena Richardson, Credit Trading Strategist at NatWest Markets Plc, notes: "What I like about AI-Powered Document Insights is that it provides a great summary of a company's comments when asked a direct question.

"For example, within an earnings call transcript for an original equipment manufacturer, the solution can be asked about the impact of tariffs. I think this solution is a godsend on a day when you have multiple earnings calls and cannot attend them all or need to revisit a topic."

Thymen Rundberg, Equity Research Associate at ING, adds: "What I find valuable is that AI-Powered Document Insights allows me to quickly find specific information within earnings call transcripts, saving me time versus manually searching through the document.

"The tool also allows me to stay up to date with more peers of companies in my coverage. This was definitely a more tedious task before."

Bloomberg plans to expand the tool with additional document coverage, features, and integrations, and has opened a beta program for future research solutions enhanced by generative AI.

Key Features

Document Insights offers natural language queries for financial transcripts through an interface that includes an 'Ask a Question' feature. Users can explore their investment hypothesis on themes such as tariffs or retrieve specific info about data points like procurement costs. 

The tool covers company documents and transcripts from various corporate events including:

• Earnings calls
• Conference presentations
• Investor days
• Capital markets days
• Shareholder meetings
• Merger and acquisition calls
• Sales results calls
• Guidance calls

The solution incorporates transparency links to highlight relevant excerpts in original documents and audio replays to capture the vocal tone of executives. It also offers 'Key Notes', providing a structured overview of topics Bloomberg has selected from company presentations.

Suzanne Szur, Research and Companies Product Manager at Bloomberg, explains: "With AI-Powered Document Insights, we're delivering our customers a solution that is infused with financial domain knowledge.

"It's been developed in close collaboration with our customers and evaluated by our team of product, data and technology experts who understand the importance of responsible AI down to the design features.

"Very soon we'll launch additional research solutions to help financial professionals and their teams find unique insights, improve their research methods and develop investment strategies tailored to various time horizons in this uncertain macro environment."

Explore the latest edition of FinTech Magazine and be part of the conversation at our global conference series, FinTech LIVE. 

Discover all our upcoming events and secure your tickets today.

FinTech Magazine is a BizClik brand