How Morningstar built a trustworthy, AI-driven financial data platform with Weaviate
Morningstar chose Weaviate's vector database to build its highly accurate and scalable Intelligence Engine Platform, offering users reliable, AI-driven financial research tools and transparent, dynamic document search capabilities.
Over the last 40 years, Morningstar amassed an extensive collection of proprietary financial data. Morningstar has sought to further empower investors that rely on its data by developing an advanced research assistant AI application. According to Benjamin Barrett, Morningstar’s Head of Technology, Research Products, building a chatbot on that data “looks like magic, but when you start peeling back the layers of the onion, you have to ask, is it actually accurate? Is it pulling the latest, greatest, most relevant data? Are our answers robust and complete?”
As his engineering team worked with Morningstar’s Quantitative Research team to build their Intelligence Engine Platform, they had to ensure an incredibly high level of accuracy in order to maintain their users’ trust. “We want to have one single source of truth. Our whole mission is to empower investor success. And the way to do that is to give them reliable, trustworthy financial data.”
In early 2023, Morningstar saw early success in experiments with LLMs and snippets of their own data. They quickly realized the potential of using AI to harness decades of longform research content and real-time data with RAG, and started their search for the right vector database.
Why Morningstar chose Weaviate:
With Weaviate, Morningstar was able to build their Intelligence Engine Platform, a product which solves a challenge facing today’s financial services firms: how to easily create and customize AI applications built on a foundation of trusted financial data and research. The Intelligence Engine also powers a variety of workflows, APIs, and chat interfaces across Morningstar’s own product ecosystem – including Mo, it’s investment research-assistant chatbot – for both internal and external users.
The Morningstar Intelligence Engine has allowed hundreds of applications to be created in house that are powering internal use cases and external workloads across their diverse portfolio of products.
Morningstar was able to launch their Weaviate-powered investment research assistant, Mo, within weeks to empower both financial professionals and individual investors to conduct investment research with greater ease.
Dynamic, context-aware document chunking and cited source transparency improves the relevance, accuracy, and trustworthiness of AI-generated answers.
Internal users can create their own applications by building a corpus of documents and using a chat interface to interact with that information.
“Through our Corpus API connected to Weaviate, users can build very powerful, low latency search engines in minutes with little to no code. Users can then also test different search algorithms without having to worry about re-indexing their data or that infrastructure at all.”
Aisis Julian
Senior Software Engineer, MorningstarMorningstar, Inc. is a leading global provider of independent investment insights. Morningstar offers an extensive line of products and services for individual investors, financial advisors, asset managers and owners, retirement plan providers and sponsors, and institutional investors in the debt and private capital markets. Morningstar provides data and research insights on a wide range of investment offerings, including managed investment products, publicly listed companies, private capital markets, debt securities, and real-time global market data.
