Tavant’s Matthew Wood: Gen AI’s New Ways of Creating How, When and Where Content is Created, Used, Required and Analyzed
Matthew Wood is Senior Director of Digital Lending with Tavant, Santa Clara, Calif.
Let it be known, Artificial Intelligence–AI–is changing how we operate our companies, the way we create products and the means of delivering services.
The transformative force of AI is so powerful that it is currently and will continue to recast entire value chains. Integrating Generative AI (Gen AI) into our products, services and operational processes affords us opportunities to create many new types of value. So I ask, how will you be using AI tomorrow? How will you be using Gen AI the day after you read this article?
Gen AI has the potential to impact every industry. Fintech, and particularly the mortgage sector, is witnessing a paradigm shift driven by various types of AI, including Machine Learning (ML), chatbots and automation. As with past technology inevitabilities (internet, smart phone, etc.), it is not whether to adopt AI but which specific applications to embrace. To get a better idea of the types of AI making waves in the mortgage industry, I went to the source–I asked ChatGPT itself.
Matt Wood: Hi, ChatGPT! What are the types of AI within the mortgage industry?
ChatGPT: Hello! In the mortgage industry, AI is increasingly being utilized to streamline processes, enhance decision-making and improve overall efficiency. Here are some key types of AI applications within the mortgage industry:
Automated Underwriting Systems (AUS):
AUS uses AI algorithms to analyze loan applications, credit scores and financial data to determine the risk associated with a mortgage application. It helps in automating the underwriting process, making it faster and more consistent.
Chatbots and Virtual Assistants:
AI-powered chatbots and virtual assistants are employed to handle customer queries, guide them through the mortgage application process and provide information on loan products. These systems can offer real-time assistance and improve customer engagement.
Predictive Analytics for Credit Scoring:
AI and machine learning algorithms analyze vast datasets to predict creditworthiness more accurately. This can result in better credit scoring models, allowing lenders to assess risks more precisely and make informed lending decisions.
Document Verification and Extraction:
AI is used for document processing, verification and extraction. Optical Character Recognition (OCR) and natural language processing (NLP) technologies help in extracting relevant information from documents, reducing the need for manual data entry.
Fraud Detection and Prevention:
AI algorithms can analyze patterns and anomalies in data to detect potential fraud in mortgage applications. This includes identifying inconsistencies in financial information or detecting unusual behavior that may indicate fraudulent activity.
Robotic Process Automation (RPA):
RPA involves using software robots to automate repetitive and rule-based tasks, such as data entry and validation. In the mortgage industry, RPA can help streamline workflows, reduce errors and enhance operational efficiency.
Personalized Marketing and Customer Relationship Management (CRM):
AI is employed for analyzing customer data to personalize marketing efforts and improve customer relationship management. This can include targeted advertising, personalized communication and customized loan offerings based on individual customer profiles.
Marketplace Lending Platforms:
AI-powered lending platforms connect borrowers with lenders in a more efficient and automated way. These platforms often use AI to assess borrower risk, determine loan terms and match borrowers with suitable lenders.
ChatGPT gave us pretty good answers. Categorically, the answers are clear and come with specific examples. So, what is missing and where are the blind spots in this response? Missing is insight into how large language models (the core of Gen AI) will change every point in our internal value chains where content is created, used, required and analyzed. From a CEO who must analyze and make data-driven decisions to the loan officer trying to respond in near-real time to real people with real questions, to a loan processor needing to educate and clarify while speeding the process forward to an underwriter whose remit is analysis and judgment rather than loan data assembly. GenAI is changing both the what and the how of work.
To give ChatGPT a try, click here.
AI does more than clever tricks with words and sentences–it eliminates mundane tasks, including 10 to 20% of processor tasks, creates valuable insights and elevates decision-making. It has the potential to revolutionize the underwriting process, significantly improving decision-making timelines and increasing throughput. AI allows C-suite executives overseeing mortgage operations to make informed decisions by providing data-driven insights. Soon, it will be the quality of our questions, not the assembling of data and loan files, that will generate the greatest value. The inevitability of this change is to understand both the value and risk and proactively embrace the new skills and capabilities that Gen AI brings.
Consider content across the mortgage industry. Whether it is as a borrower, a seller, a prompt in a workflow, the raw information of a mortgage, what to say in an email or chats that are actually useful, Gen AI will play a role in all of these and much more. It will help your company pioneer automation and insights that are uniquely yours, leading to a future where operations are not just different but remarkably efficient, humane and forward-thinking. The window of opportunity is open.
As for me, Gen AI is going to take my meeting’s notes, help my software developers test their code, extract data from documents and run ambient propensity analysis. How about you? How will you use Gen AI tomorrow?
(Views expressed in this article do not necessarily reflect policies of the Mortgage Bankers Association, nor do they connote an MBA endorsement of a specific company, product or service. MBA NewsLink welcomes your submissions. Inquiries can be sent to Editor Michael Tucker or Editorial Manager Anneliese Mahoney.)
