Artificial intelligence (AI) is a promising and much discussed technology that may change how we work in ways we can’t even imagine. Will this new technology transform commercial mortgages and eliminate jobs, as some are predicting? Or is this just another fictional wizard?
We don’t know the answers to these questions just yet, but evolution appears to be occurring quickly. There has already been a frenzy of media attention surrounding AI that has helped students write academic papers, has created photos and songs that mimic the styles of famous artists, and has the ability to answer your questions in a conversational format.
“AI systems will raise concerns about privacy, along with the potential for the data to be misused or shared without consent.”
To better understand the real-world capabilities of this next-generation technology, ChatGPT was consulted with the following question: Is AI the real thing or a wizard behind a curtain? The system took some time to generate a response, but that may have been because the free version was used.
Here is ChatGPT’s answer: “The idea of AI being a wizard behind a curtain is a metaphorical reference to the Wizard of Oz, a powerful being hiding behind a curtain and manipulating events from behind the scenes. While AI can certainly be a powerful tool that can impact many aspects of our lives, it is not inherently mysterious or manipulative. AI is a field of computer science that focuses on developing algorithms and systems that can perform tasks that typically require human-like intelligence, such as recognizing speech, images, or patterns in data.
“AI systems are designed and developed by teams of engineers, researchers and data scientists who work to create systems that can learn, adapt and make decisions based on data and experience. While AI has the potential to transform many industries and aspects of our lives, it is ultimately a tool that is created and controlled by humans. It is not an all-powerful being that can control events from behind the scenes.”
A powerful tool
With the potential for AI to be a powerful new tool, how will the commercial mortgage and real estate industries use it? History may provide some answers.
More than 60 years ago, a crude form of a new technology was pioneered that allowed users to organize information, perform calculations and analyze data in a computerized table format. It might be hard to remember that far back, but when spreadsheet applications were introduced, many wondered if their jobs would become obsolete.
Rather than eliminating jobs, spreadsheets automated repetitive tasks, performed complex financial analyses and empowered data-driven decisions. Spreadsheets are commonly used in the commercial mortgage process and across many other business tasks. The technology may have eliminated some low-level jobs, but it also freed workers from the drudgery of data-intensive and time-consuming tasks.
In recent years, the potential of big data captivated the technology industry with its ability to create large and complex data sets from various sources, including social media, scientific research and business transactions. One of the main promises of big data is that it can be analyzed to reveal insights, trends and patterns that were not previously apparent.
By using AI technology, these analyses are now possible through machine learning algorithms, which can recognize patterns across various media such as text, images and videos. Once patterns are recognized, AI systems can make rule-based decisions based on the data. Combined with natural language processing technology, AI can interpret, understand and generate human language.
For the commercial mortgage industry, this has profound implications. AI has the potential to automate the analysis of narrative text in loan documents, thereby saving valuable time that people can allocate to more complex and strategic tasks. This technological advancement can make the industry more efficient and facilitate higher-level decisionmaking.
On the market
There are companies using AI technologies in commercial real estate today. To enable collaboration during the construction process, OpenSpace uses 360-degree cameras and computer vision to create digital replicas of construction sites. These videos capture a record of the construction project, allowing teams to remotely monitor and document progress. This saves time and money for travel and prevents budget overruns due to change orders.
CompStak uses AI technology to collect and analyze large amounts of data from a variety of sources (leasing brokers, property owners, public land records, etc.) and subsequently provide analyses of commercial properties, leases and sales. By identifying patterns and trends in the data, users can have deeper insights about potential risks and opportunities.
Similarly, Skyline AI (which was acquired nearly two years ago by JLL) is a platform that analyzes commercial property data to predict investment opportunities. The system collects and analyzes data from financial statements, tenant information, property values and other sources. The platform then utilizes machine learning algorithms to identify properties that are likely to generate strong returns, thereby enabling investors to make well-informed decisions.
Future transformation
How might AI transform the industry in the future? Digitizing loans is a challenge today as a commercial mortgage deal can include hundreds of pages of documents. Sorting, organizing and maintaining different versions of these documents are manual tasks.
AI could be used to automatically classify and sort loan documents by type, such as tax returns, rent rolls, financial statements and correspondence. Natural language processing may help accelerate the digital transformation of the data and documents by analyzing and identifying common patterns, structures and relationships in existing loan documents.
Key data, such as loan amounts, interest rates, repayment terms and collateral information, could be extracted in a standardized manner. Once extracted, the data could be checked for inconsistencies across the various documents. Additionally, by analyzing the language in the loan agreements referenced in the promissory note, AI could identify potential inconsistencies with representations and warranties, default provisions, lease expirations and payment schedules.
The right data
For AI to be accurate, it needs a large and diverse set of data that is representative of the types of necessary documents. In commercial real estate, this would include loan applications, financial statements, appraisal reports, sales agreements and more. These documents also have a varied narrative language.
For instance, a promissory note might contain the language, “For value received, the undersigned (whether one or more, “borrower”) promises to pay to XYZ Bank (“bank”) the sum of three million and no/100 dollars ($3,000,000.00).” Another promissory note might read, “In return for a loan that I have received, I promise to pay U.S. $3,000,000.00 (this amount called “principal”) plus interest, to the lender, XYZ Bank.” After analyzing the documents and accounting for the variability in the narrative text, AI could extract key pieces of information and other relevant details, such as in the example above: the lender’s name as “XYZ Bank” and the loan amount as “$3,000,000.”
To accelerate the digitization of commercial mortgages, AI could analyze documents to generate uniform templates for different types of documents, such as loan agreements, promissory notes and security agreements. With the import of standardized data, AI algorithms could automatically prepare loan documents by having previously identified where data is in the text. The algorithms could be customized for borrower- or loan-specific information.
A borrower’s financial information could be used to generate specific loan covenants or clauses. Machine learning rules may include specific regulatory or legal requirements such as zoning restrictions, environmental regulations and land-use laws. Using AI technology to translate loan data and documents could increase efficiency, reduce costs and improve accuracy.
Future challenges
Introducing a new technology has potential benefits but also brings challenges. AI systems will raise concerns about privacy, along with the potential for the data to be misused or shared without consent.
As with any technology, AI systems are vulnerable to cyberattacks that could further compromise sensitive information. AI needs safeguards to ensure that the technology is used in ethical and responsible ways. AI systems are only as good as the data on which they are trained. So, if the data is biased or incomplete, the results generated by the AI will also be biased or incomplete.
Additionally, users need to understand that AI algorithms may produce unexpected or undesired results. It is important for technology solution providers and lenders to be alerted to these potential issues and take steps to mitigate them. A review by humans, not robots, will be necessary at every stage of development and use.
● ● ●
It’s going to take time, but AI may become a revolutionary business tool that will automate repetitive tasks while enabling faster and more accurate decisionmaking. We are off to see the wizard, but this may be a very long yellow brick road to the Emerald City.
This journey will take the commercial mortgage industry to some new and unfamiliar places that may even be a little scary. There are challenges to using AI, but there also are many potential automations and outcomes that have yet to be imagined. Toto, it feels like we’re not in Kansas anymore. ●
