Better.com Teams Up With OpenAI to Embed Its Tinman Credit Engine Inside ChatGPT, Offering Real‑Time Mortgage Decisions
The partnership introduces a custom Model Context Protocol (MCP) connector that bridges Better’s Tinman AI Platform with ChatGPT’s large‑language model environment. Unlike a simple API call, the MCP supplies the model with a live snapshot of every loan file—facts, documents, pending actions, and required work—so that each request is answered with the precise context needed for underwriting. The result, according to Better, is a decision tree that mirrors the depth of traditional loan‑origination systems while cutting processing time dramatically.
Speed that challenges industry norms
In a demonstration, Better’s Tinman app was able to ingest a borrower’s application data, parse supporting documents, apply institutional pricing and underwriting guidelines, and output a complete credit decision in as little as 47 seconds. The median processing time recorded was 2 minutes 24 seconds, a stark contrast to the 21‑day average for conventional mortgage pipelines. The speed gains stem from Tinman’s real‑time context graph, which tracks over $110 billion in funded loans, 12 million recorded customer calls, and more than 5 billion pages of credit, income, and asset documentation.
Why the move matters for lenders
The mortgage sector has long been criticized for its reliance on legacy loan‑origination systems (LOS) that are fragmented, slow, and costly. By embedding Tinman’s decision engine inside a platform that already enjoys a user base of over 100 million Americans, Better aims to lower operational overhead for banks, fintechs, and independent loan officers. “The mortgage industry is riddled with inefficiencies that hurt consumers, as well as the loan officers and lenders who serve them,” said Leah Price, General Manager of the Tinman AI Platform. “Loan officer teams and banks can simply log into their ChatGPT Enterprise account, download the Tinman AI credit decision engine app, connect their guidelines, pricing, and CRM to process, underwrite, and fulfill loans nearly instantly; passing thousands of dollars in savings to consumers.”
A broader AI strategy for Better.com
The Tinman integration builds on Better’s enterprise‑wide rollout of ChatGPT across its 1,400‑person workforce, where the company already uses OpenAI’s multimodal models to power Betsy™, an AI technology loan agent that assists both consumers and enterprise clients. “Better is embedding AI at the core of its business, helping its employees and lenders move faster from data to decision,” noted Giancarlo “GC” Lionetti, Chief Commercial Officer at OpenAI. “With OpenAI, Better is not only advancing mortgage intelligence for the industry but also demonstrating how AI can transform how financial institutions operate from the inside out.”
Competitive positioning in a crowded fintech landscape
While several fintechs have experimented with AI‑enhanced underwriting, Better’s approach is distinguished by the depth of its proprietary data set and the breadth of its institutional relationships. Tinman’s context graph incorporates underwriting criteria and pricing parameters from more than 45 institutional buyers of mortgage and home‑equity loans, collectively representing over 80 % of the U.S. mortgage market. This coverage includes government‑sponsored enterprises such as Fannie Mae and Freddie Mac, FHA loans, and major banks like J.P. Morgan Chase, Truist, Citizens Bank, US Bank, Huntington Bank, and Fifth Third.
Real‑world feedback from the field
Tony Song, a Better loan officer who ranks in the Top 100 for dollar volume, purchase volume, and loan count according to Scotsman Guide, praised the new tool: “The most important part of my day is delivering mortgage purchase commitment letters to homebuyers that need a mortgage to buy a home. With this launch, I’m already benefitting from the efficiencies, and I will be able to serve 10x more customers daily compared to what was possible with the traditional mortgage underwriting process.” Such testimonials suggest that the technology could reshape daily workflows for loan officers, shifting the bottleneck from manual document review to strategic client interaction.
Potential regulatory implications
Embedding a credit decision engine within a conversational AI raises questions about compliance with fair‑lending regulations and the Equal Credit Opportunity Act (ECOA). Better’s Tinman platform claims to retain a “granular decision tree” that logs the rationale behind each approval or denial, a feature that could satisfy audit requirements. However, regulators may scrutinize how the AI model interprets protected class data and whether the MCP’s real‑time context could inadvertently expose sensitive borrower information. The company has not disclosed any formal regulatory review, but its emphasis on “decision‑ready information” suggests an awareness of compliance needs.
Market impact and future roadmap
The Tinman‑ChatGPT integration signals a broader trend of fintechs leveraging large‑language‑model ecosystems to deliver point‑of‑sale intelligence. By moving underwriting from a back‑office function to a conversational front‑end, Better positions itself as a provider of “embedded finance” capabilities that could be licensed to other financial institutions. Vishal Garg, Founder and CEO of Better, emphasized the strategic intent: “We have full intention to continue building additional tools on OpenAI that will revolutionize the mortgage and home equity industry together. By reducing the time and cost to originate, we can dramatically lower rates for consumers and make mortgages more affordable for more American families. We’re just getting started.”
How to experience the demo
Better has made a public demo of the Tinman AI Credit Decision App available at https://better.com/tinman. The demo walks users through a typical mortgage application, showing how the AI parses documents, applies institutional guidelines, and returns a decision within seconds. While the demo is not a production system, it offers a concrete look at the workflow that lenders could soon adopt.
Outlook for AI‑driven lending
The collaboration between Better and OpenAI underscores how large‑language‑model platforms are evolving from chat interfaces to transactional workhorses. As more financial institutions seek to digitize and accelerate credit decisions, the combination of a deep, industry‑specific data set (Tinman) with a flexible conversational layer (ChatGPT) could become a template for other verticals—auto loans, small‑business credit, and beyond. The key challenge will be balancing speed with transparency, ensuring that AI‑generated decisions remain auditable and compliant with existing lending laws.
Get in touch with our fintech expert
