Last month, MISMO, the mortgage industry's standardisation organisation, introduced two new certifications for AI governance. These are designed to provide mortgage companies, lenders, and asset managers with a formal standard for evaluating automated systems. Building on the organisation's framework for responsible AI in the mortgage ecosystem, the certifications establish explicit benchmarks for risk monitoring, influence on decisions, and the documentation of source data.
This announcement marks a permanent shift in how the commercial property industry assesses technology. The discussion is moving from operational convenience to institutional compliance. Over the past two years, real estate technology focused almost exclusively on content generation. Tools summarised leases, created investment memos, and performed cash flow projections in record time. However, when these automated workflows reached institutional credit committees, a clear limitation emerged.
Requirements for Traceability
Lenders and limited partners quickly realised that an attractively formatted model is useless if no one can prove which file version generated the net operating income figure, or how a non-standard debt agreement was interpreted. The introduction of formal governance frameworks signals that capital providers will no longer accept this lack of transparency. Over the next 12 to 18 months, this shift will create a clear division within the property capital markets: an auditability spread.
Companies utilising certified, deterministic software architectures will experience faster deal closures, smoother syndications, and reduced transaction friction. Conversely, deal teams relying on unconfirmed black-box results will have to endure longer credit reviews, more elaborate manual audits, and higher capital costs. If institutional lenders cannot trace data back to its source, they treat this gap as a financial risk.
The Future Belongs to Domain-Specific Architectures
This dynamic changes how operators must structure their internal technology. Previously, procurement teams routinely purchased general AI tools, expecting them to handle complex property logic. This strategy largely failed, as general language models lack the domain rules and historical context of an institutional firm. The future belongs to domain-specific architectures that break down underwriting into narrow, independently auditable steps.
- —A micro-process tracks document version control.
- —Another verifies the integrity of spreadsheet formulas.
- —A third reconciles extraction results with general ledgers.
When each step generates a clear audit trail, the end result becomes traceable work rather than a high-tech guess. With the establishment of these standards, senior property managers must rethink their internal review processes. Junior analysts will continue to use software to skip manual data entry, but executives must use the time gained to question the underlying assumptions behind every result. The era of trusting software because the presentation looks clean is officially over.
Real estate finance is returning to a fundamental truth: if you cannot verify the figures, you do not own the asset – you own the risk. Arunabh Dastidar is co-founder and CEO of the real estate investment platform Leni.
