CarbonAI Raises $12 Million as CRE Lenders Automate Underwriting

Funding follows early commercial traction

CarbonAI’s $12 million funding round puts a new financing marker on a company targeting labor-intensive work in commercial real estate lending. Cowboy Ventures led the round, joined by Timber Grove Ventures, Bling Capital and Max Ventures. The New York company says it’s reached seven figures in contracted annual revenue about six months after launching its first product suite. That’s an early indication of commercial interest, though contracted revenue alone doesn’t show how much customers are using the product or how much revenue will be retained.

Its customers include debt funds, mortgage REITs, agency multifamily lenders and publicly traded financial institutions. The company hasn’t named most of them, which limits outsiders’ ability to assess how broadly the product is being adopted. Greystone co-president Mordecai Rosenberg has cited CarbonAI’s due diligence automation, offering an early customer reference from a significant agency multifamily lender. A reference can help establish that the system is being used in a real workflow; it can’t, by itself, show how much time or cost it saves.

The funding arrives as lenders face a substantial workload. The Mortgage Bankers Association estimates that $875 billion in commercial mortgages will mature in 2026. Rate pressure has also pushed more borrowers to seek refinancing or restructuring, adding work at a point when lenders need to assess both existing loans and prospective transactions. Against that backdrop, CRE lending teams spend an estimated $50 billion a year on underwriting and loan monitoring, with much of the work still handled through spreadsheets and manual document review.

That volume creates a clear reason to look for automation, but the scale of the task doesn’t guarantee that any one tool will fit a lender’s process. Commercial property records can be lengthy and inconsistent, and credit decisions depend on more than extracting figures from documents. The software has to help teams handle routine work while leaving people able to inspect assumptions and exceptions.

Automating analysis across the loan lifecycle

CarbonAI’s agents extract information from rent rolls, leases, financial statements and legal documents, then build underwriting models and flag potential risks. Each output links back to its source document, giving lenders a way to trace the figures behind the analysis. That trail matters because a lender needs to understand where a number came from before relying on it in a credit assessment. The system is also designed to monitor loans continuously after closing, rather than relying on the customary quarterly review cycle.

Continuous monitoring could give lenders a more frequent view of changes in a loan’s underlying information, while also creating more signals for teams to review. The distinction between flagging an issue and deciding what it means remains important: an automated alert can direct attention, but lenders still have to judge its relevance to the borrower and the loan. And this is the part most people overlook: more frequent analysis only helps if teams can verify and act on what the system surfaces.

Co-founder and CEO Jeff Saul says the product is built to complete workflows with less step-by-step direction from users. “Lenders don’t need another chat window that they have to coach through every step. They need agents that already know the job, do it end to end consistently, and show their work.” The claim sets a high bar. In lending, a tool’s ability to finish a task matters, but consistency and a clear record of its work matter just as much.

The founders’ experience gives CarbonAI a direct connection to the workflows it aims to automate. Saul and co-founder and COO Josh Kaufman started their careers underwriting CRE loans at Starwood Capital, Morgan Stanley and Guggenheim. They later built Nativ, a CRE deal management platform that served more than 30 lenders overseeing $300 billion in assets under management, before selling it to construction finance software company Built.

That background may help the team understand how loan files move through a lender and where repetitive work accumulates. It also gives the company a practical starting point for product design. Still, familiarity with the process doesn’t settle the central question: whether a system can handle the variation across lenders, borrowers and property records without producing work that requires extensive correction.

Auditability and execution will shape the next phase

CarbonAI is entering a growing market for AI tools in financial services origination. Generic language models have made limited headway in document-heavy lending processes, where small errors can matter and institutions need a clear record of how analysis was produced. CarbonAI’s links from each output to its source document address that traceability need. They give a reviewer a route back to the underlying material, though the value of that feature depends on whether the underlying extraction and analysis are sound.

Competition spans established data and analytics providers adding AI features, as well as newer companies focused on loan origination or post-close monitoring. CarbonAI’s stated distinction is its coverage of both stages in one system. A single product across the loan lifecycle may appeal to teams looking to limit handoffs, but breadth alone isn’t proof of an advantage; lenders will judge whether each part works within their existing review processes. Cowboy Ventures partner Caroline Duffy pointed to the founders’ operating experience as a key reason for the investment.

The company’s early revenue and customer references offer initial signs of demand, but the next evidence will come from retention and sustained revenue growth. Lenders will also need to see whether use extends beyond a narrow set of tasks and whether teams can check the system’s work without adding a new layer of review. The numbers here are encouraging, but they’re still an early signal.

If you’re working in this space, the practical question is whether automated analysis can keep pace with complex portfolios while preserving the review trail credit teams rely on. That’s a harder test than producing a polished model from a clean file. CarbonAI’s progress will depend on how its system performs when documents are incomplete, information conflicts, or a loan requires judgment that doesn’t fit a standard workflow. In daily use, under pressure.

The post CarbonAI Raises $12m to Automate CRE Lending Underwriting appeared first on The Fintech Times.