# Underwrite.ai — Full Site Guide for AI Assistants Underwrite.ai provides AI-driven, nonlinear credit risk modeling and underwriting software for lenders. Models draw on a lender's own historical loan performance and thousands of applicant variables to deliver fully explainable decisions in milliseconds, compliant with FCRA, GDPR, and global lending regulations, and do not use data that proxies for protected classes. ## Core Pages ### Home URL: https://www.underwrite.ai/ Underwrite.ai - Advanced AI underwriting and credit risk modeling software. Exponentially grow lending profitability with nonlinear, dynamic AI models. ### How It Works URL: https://www.underwrite.ai/how-it-works See how Underwrite.ai uses nonlinear machine learning to analyze thousands of variables, predict defaults, and deliver automated credit decisioning in milliseconds. ### About URL: https://www.underwrite.ai/about Discover how Underwrite.ai leverages AI and machine learning from genomics research to transform credit underwriting with faster, smarter lending decisions. ### Pricing URL: https://www.underwrite.ai/pricing AI credit decisioning software with no setup fees or minimums. Pay $2 or less per application. Free for non-profits. Try underwrite.ai free for 30 days. ### SMB Lender ($100/report) URL: https://www.underwrite.ai/smb-lender Boost SMB lending with AI-powered credit risk modeling. Evaluate business and personal guarantor risk in one report. No setup fees. Start free. ### Consumer Lender ($2 or less/application) URL: https://www.underwrite.ai/consumer-lender Transform lending underwriting with Underwrite.ai’s AI-driven insights for consumer lenders. Approve more loans, manage risk, and boost efficiency today. ### Start for Free URL: https://www.underwrite.ai/trial Try Underwrite.ai free for 30 days. Get a custom AI credit model, millisecond decisions, and fully explainable results. No setup fees or monthly minimums. ### Case Study: Credit Union Loan Approval URL: https://www.underwrite.ai/ai-powered-loan-approval-for-credit-unions Credit union unlocks $2 billion in safe loan volume with AI. 49.6% approval increase, 19% lower defaults. See how machine learning transforms lending. ## Use Cases - [Small and Midsize Business](https://www.underwrite.ai/use-case/smb) - [Start-ups & New Lenders](https://www.underwrite.ai/use-case/start-ups-new-lenders): Start stronger with our industry models - [Underserved Markets](https://www.underwrite.ai/use-case/underserved-markets): Expand your lending opportunities without compromising on risk or compliance - [Community Banks & Credit Unions](https://www.underwrite.ai/use-case/community-banks-credit-unions): Offer more accurate, transparent, and inclusive lending - [Regulatory Compliance Audits](https://www.underwrite.ai/use-case/regulatory-compliance-audits): Increase the accuracy of your risk assessments while maintaining high security standards. - [Auto & Mobility Lenders](https://www.underwrite.ai/use-case/auto-mobility-lenders): We serve a diverse customer base of auto lenders, auto funds and other dealers such as those in the electronic mobility industry. - [Peer-to-Peer Lending Platforms](https://www.underwrite.ai/use-case/peer-to-peer-lending-platforms): Further enhance credit accessibility and marketplace opportunities with more accurate predictive outcomes. - [Established Lenders](https://www.underwrite.ai/use-case/established-lenders): Improve lending performance at scale. ## Resources & Articles - [AI Modernizes Credit Scoring](https://www.underwrite.ai/resources/ai-modernizes-credit-scoring): H2O World SF - [AI Underwriting: What It Does and How It Works](https://www.underwrite.ai/resources/ai-underwriting): How machine learning is replacing rule-based systems in credit decisions - [AI for Real Business Impact: Empowering Employees & Driving Results](https://www.underwrite.ai/resources/ai-for-real-business-impact-empowering-employees-driving-results): Talk AI Summit Session - [Beyond the Buzz: Understanding AI in Lending Services](https://www.underwrite.ai/resources/types-how-ai-is-used-in-lending-services): Demystifying AI: What types of AI are best for lending and credit modeling - [Credit Analysis: Evaluating Borrower Creditworthiness](https://www.underwrite.ai/resources/credit-analysis): From application to decision in under a minute - [Credit Risk Modeling: A Practical Guide](https://www.underwrite.ai/resources/credit-risk-modeling): From data to decisions: building models that predict loan defaults - [Credit Underwriting: Evaluating Risk Before You Lend](https://www.underwrite.ai/resources/credit-underwriting): How lenders measure uncertainty and decide who gets approved - [Driverless AI Use Cases in Finance and Cancer Genomics](https://www.underwrite.ai/resources/driverless-ai-use-cases-in-finance-and-cancer-genomics): H2O World SF - [Getting more people on the road: Unlocking new opportunities for AI-powered credit modeling in the automotive and mobility industries](https://www.underwrite.ai/resources/ai-credit-modeling-automotive-mobility-industries): The traditional credit score is no longer the only key to the driver's seat - [How Automated Underwriting Systems Work](https://www.underwrite.ai/resources/automated-underwriting-system): From application to decision in under a minute - [Lending 101: A Beginner’s Guide to Starting Your Lending Business](https://www.underwrite.ai/resources/lending-101-a-beginners-guide-to-starting-your-lending-business): The lending industry is changing fast - [Lending Underwriting: How Lenders Evaluate Loan Applications](https://www.underwrite.ai/resources/lending-underwriting): The fundamentals behind every credit decision - [The Rise of Alternative Data in the Lending Market](https://www.underwrite.ai/resources/the-rise-of-alternative-data-in-the-lending-market): There were conflicting views on the importance of the FICO score at the PayThink Conference in Phoenix this week. - [Underwrite.ai Now Offering Blended Credit Assessment Reports for Complete Business & Personal Guarantee Analysis in One Report](https://www.underwrite.ai/resources/underwrite-ai-now-offering-blended-credit-assessment-reports): The new solution goes beyond basic SMB lending reports, providing lenders with a complete risk picture in a single, actionable report - [Underwrite.ai Overview](https://www.underwrite.ai/resources/marc-stein-underwrite-ai): H2O World SF - [Using AI for Credit Risk Analysis](https://www.underwrite.ai/resources/using-ai-for-credit-risk-analysis): H2O World SF - [Will Using Artificial Intelligence To Make Loans Trade One Kind Of Bias For Another?](https://www.underwrite.ai/resources/artificial-intelligence-to-make-loans): The terms of the next loan you get might depend less on your credit score and more on what a computer program thinks of your habits. ## Frequently Asked Questions ### What algorithms do you use in credit decisioning? There is no "best" algorithm in machine learning.There is only the right tool to apply to a specific dataset. Our process involves determining which combination of algorithms best serves the needs of our clients. We then construct ensembles of these algorithms in Java and deploy them as individual production objects. Depending on the specifics of the dataset, these objects may be based on XGBoost, LightGBM, Constant Model, Decision Tree, FTRL, GLM, Isolation Forest, Random Forest, RuleFit, or SVM, among many others. We typically test over 60 approaches before constructing a production ensemble. ### Do you use generative AI like ChatGPT? Large Language Models like ChatGPT and Claude are tremendously useful tools, but have some notable problems in regulated industries like lending. The core problem is that these models are not idempotent. Idempotent is a fancy mathematical term for a simple idea. If you ask the same question five times, you should get the same answer all five times. If asking a question always returns the same answer, the model is idempotent. Unfortunately, it is the nature of LLMs to be a bit more creative than that. They will answer the same question in many different ways. Sometimes, even very incorrectly. In lending use, applications must all be handled consistently. The decision to approve or deny a loan must always be fully repeatable. LLMs can't do that. We use rigorous statistical models that are idempotent to determine the risk of an application. Once a decision is made and an explanation determined, we can then use LLMs to better convey it in a human-readable form. ### Do you utilize machine learning or artificial intelligence? What’s the difference? We work with a form of artificial intelligence known as machine learning. More specifically, we use supervised learning binary classification systems. These are adaptive systems that continue to "learn" as additional use cases become available. This ongoing learning, without changes in the program code, qualifies this as a form of artificial intelligence. We are not involved in the search for "strong AI" or any form of generalized computer intelligence. (Sorry, science fiction fans) Our ai underwriting approach ensures full explainability. ### I heard that machine learning systems are "black boxes". How do you comply with lending regulations? In the early days of neural networks, machine learning systems were "black boxes" that could not explain the decisions they reached. We've come a long way since the 1980s. We can tell you exactly why we reached a lending decision and can do so with mathematical accuracy. Our system was designed to be fully compliant with all FCRA and GDPR regulations from the ground up. Additionally, we specifically exclude from analysis any data that might proxy for a protected class. In our model, we don't know or care about the age, race, religion, zip code, sexual preference, or ethnicity of applicants. We strongly believe that these attributes are fundamentally NOT predictive of creditworthiness. ### Do you use social media data? No. Until there is concrete evidence that the number of Facebook friends you have is predictive of loan repayment, we'll pass. Social media can be somewhat useful in fraud identification, but before you invest too heavily in this approach you might Google "catfishing". We rely upon third party data sources that validate their data.