The real AI divide in the mortgage industry

Despite substantial investment in artificial intelligence, mortgage origination remains expensive and labor-intensive. In the first quarter of 2026, the average cost to originate a loan rose to $11,898, while lenders earned just $727, or 16 basis points, in pre-tax production profit. Why has that investment not produced a visible reduction in the cost to originate? The answer is structural. Most mortgage AI deployments improve individual tasks inside the existing workflow. The larger opportunity is to redesign how mortgage companies actually operate. Why mortgage lending remains a human-capital industry Mortgage remains a human-capital industry. Across origination, underwriting, closing, secondary-market execution and servicing, growth still requires more people to collect and verify information, resolve exceptions and manage handoffs across systems and institutions. Small firms add people directly to the workflow. Large firms add production capacity and organizational layers to manage it. The company becomes larger without necessarily becoming structurally more profitable. That reliance on people is a response to severe process fragmentation. Mortgage lending is combinatorially co...