Artificial intelligence (AI) is no longer a futuristic concept; it’s a present-day reality that is reshaping industries across the board, including the financial services sector. Among the many sectors embracing this technological wave is commercial real estate (CRE). 

With this fast development in the industry, Smart Capital Center hosted a Lunch and Learn during the MBA Servicing Conference to share the disruptive potential of generative AI in CRE servicing and asset management. 

In this article, you will: 

•  Gain insights into the current state of generative AI and its implications for the CRE finance     
   industry.

•  Explore the transformative applications of this technology across the loan lifecycle.  

•  Understand the challenges and considerations surrounding the adoption of generative AI models. 

• Get a glimpse into the future of AI in CRE lending and the steps companies can take to prepare. 

Generative AI and its Current State 

Generative AI has demonstrated significant benefits at the individual level. As Laura Krashakova highlighted, the benefits of leveraging this technology are substantial. A research made by Boston Consulting Group revealed that consultants using AI “finished 12% more tasks, completed tasks 25% quicker, and produced 40% higher quality results than those without.” 

According to the Times, AI has surpassed humans in numerous tasks such as handwriting recognition, speech and image recognition, reading comprehension, and language understanding, and this trend of AI outperforming human capabilities is accelerating. 

The Disruptive Potential of Generative AI in CRE 

Generative AI has the potential to transform commercial real estate (CRE) by acting as a versatile tool in various roles, such as a collaborative human-AI writing agent or a specialized underwriter and asset manager. This flexibility allows AI to assist in creating content, analyzing data, and managing assets more efficiently than ever before. 

In CRE lending and asset management, generative AI can be applied across the entire workflow: 

•  Ingest: Gather data from diverse formats and sources, including property documents, market reports, and financial statements. 

•  Organize: Structure and categorize the ingested data for efficient analysis. 

•  Analyze: Identify patterns, trends, and insights within the data to inform decision-making. 

•  Generate: Produce reports, summaries, visualizations, and other outputs that communicate findings and recommendations. 

Select Use Cases for CRE Lending 

 1. Summarization of Information: Generative AI can efficiently summarize a wide range of documents, including deal materials, market overviews, and complex legal documentation. As Laura points out, “Prompt engineering is key”  to obtaining accurate and relevant summaries. This involves selecting the right prompting technique, crafting detailed prompts, and potentially utilizing AI-generated prompts for optimal results. Personalization can further enhance the quality of summaries, and tools such as Perplexity.ai can be leveraged to achieve this.

2. Research: Generative AI is a powerful research tool that can quickly assess tenant risk, provide market insights, identify deal risks, and even summarize local regulations. This can significantly accelerate the research process and empower lenders and asset managers to make more informed decisions. 

3.  Extracting and Standardizing Data from Documents: AI-powered tools such as Smart Capital can extract and standardize data from various documents. This streamlines the data collection and analysis process, enabling faster and more accurate underwriting and asset management decisions. The extracted data can then be used for financial analyses, variance analyses, normalizations, and reporting. 

4. Generating Visualizations, Property, and Portfolio Insight: Generative AI can be used to generate interactive charts and visualizations.  Such visual representation aids in quickly and effectively assessing financial and operating property and borrower performance, market analysis, and the like. 

In addition to the default charts generated by underwriting systems such as Smart Capital, users can now create their own visualizations within minutes without the need for complex SQL queries and data manipulations. Instead, users can simply write a description of the chart they want to see, and the system will generate it in real time.  

This process significantly improves the efficiency and accuracy of creating comprehensive investment memos.  

5. Automating the Generation of Credit Memos, Deal Reviews, and More 

Generative AI is transforming the way lenders prepare credit memos by streamlining the traditionally manual and time-consuming process. With Generative AI, lenders can automate the initial drafts of credit memos, allowing the AI to generate key sections such as borrower summaries, financial analyses, and risk assessments. This automation not only saves time but also enhances accuracy by minimizing human errors.

6.  Optimizing Communication Between Lenders, Borrowers, and Other Stakeholders: Generative AI can optimize communication by generating better and faster notes and messages. For instance, Smart Capital New Generation Borrower Portal can create personalized alerts, reminders, and notifications for both internal deal teams and external parties. Additionally, AI can enhance the personalized borrower portal experience, making interactions more efficient and tailored to individual needs. This improved communication helps to maintain transparency and strengthen relationships between lenders, borrowers, and other stakeholders. 

The Rise of AI Agents: Transforming Workflows in CRE 

One of the most interesting innovations today is the concept of AI Agents. These are autonomous AI programs designed with specific goals. These agents can independently work toward accomplishing objectives assigned to them. 

A notable example of this innovation is Devin, the first AI software engineer developed by Cognition, launched in March 2024. Devin exemplifies the capabilities of advanced AI agents in the following ways:

•  Learning Unfamiliar Technologies: Devin can quickly learn and adapt to new technologies, making it a versatile tool for various applications.

•  Building and Deploying Apps: Devin can handle the entire process of building and deploying applications, from start to finish, without human intervention. 

•  Autonomously Fixing Bugs: Devin can identify and fix bugs within codebases, ensuring smoother and more reliable software performance. 

•  Training and Fine-Tuning AI Models: Devin is capable of training and fine-tuning its own AI models, continually improving its performance and accuracy. 

•  Addressing Bugs and Feature Requests: In open-source repositories, Devin can autonomously address bugs and feature requests, contributing to the development and maintenance of mature production repositories. 

Beyond software engineering, AI agents such as Devin can act as hundreds of specialized agents in the CRE sector, including underwriters, closers, and asset managers. These agents can autonomously perform tasks such as underwriting loans, managing assets, and closing deals, drastically scaling operations, and allowing smaller companies to compete with larger players on an equal footing. 

The emergence of AI agents will mark a new era in automation, significantly transforming workflows in commercial real estate (CRE). These intelligent agents are designed to handle complex tasks autonomously, enhancing efficiency and productivity across various roles in the industry. 

As Laura Krashakova explains, “Imagine having tens or hundreds of agents working alongside your team – underwriters, asset managers, closers, drastically scaling your operations.” 

This transformation paves the way for a more efficient, productive, and innovative future in CRE lending and asset management. 

Risks of AI: Reliability, Data Privacy, Bias, Cost, Regulation 

While the potential of AI in commercial real estate (CRE) lending and asset management is immense, it is essential to address the associated risks to ensure responsible and effective implementation. The primary concerns include: 

•  Reliability: AI, particularly generative models, can sometimes produce inaccurate or nonsensical outputs, often referred to as “hallucinations.” Ensuring the reliability and accuracy of AI-generated information is important, especially in the financial sector where decisions have significant consequences. 

•  Data Privacy: Protecting sensitive data is paramount. AI models, especially those used in financial applications, must handle data responsibly to prevent breaches and misuse. Many enterprise solutions now offer options to keep data secure, such as private instances of language models that do not share data with external sources. 

•  Bias: AI systems can inadvertently learn and perpetuate biases present in training data. This can lead to unfair or discriminatory outcomes. Ongoing efforts are required to identify and mitigate biases in AI models. Implementing diverse training datasets and rigorous testing can help reduce these biases. 

•  Cost: Although AI can reduce operational costs in the long run, the initial investment can be significant. Training sophisticated models and integrating them into existing systems require substantial financial resources. However, as AI technology continues to evolve, the costs are expected to decrease, making it more accessible. 

•  Regulation: The regulatory landscape for AI is still developing. Organizations must stay abreast of new regulations and ensure compliance to avoid legal and ethical pitfalls. Regulations may cover various aspects, from data protection to the ethical use of AI, and businesses must be prepared to adapt to these evolving requirements. 

Impact of AI on the Future of Work 

The impact of AI on the future of work is profound, with large language model (LLM) agents demonstrating superhuman performance in various tasks. These advancements bring several key changes to the workplace: 

•  AI as a Skill Leveler: AI serves as a skill leveler by boosting the performance of lower-performing employees. With AI support, all team members can achieve higher levels of productivity and quality in their work. 

•  Differentiating Employees by AI Proficiency: In the future, the ability to use AI effectively will differentiate employees. Those proficient in leveraging AI tools will be in high demand, as their skills will be crucial for maximizing the benefits of AI integration. 

•  Blending AI and Human Efforts: The most effective approach involves blending AI with human efforts. By outsourcing repetitive and data-intensive tasks to AI, and having humans focus on areas requiring judgment and creativity, organizations can achieve more accurate, varied, and high-quality results. 

To further explore AI’s evolving role in commercial real estate, check out our earlier discussion on the future role of generative AI in CRE Lending.

10

8

9

11

AI is not just a tool but a partner that enhances human capabilities. To fully understand the transformative role of generative AI in servicing and asset management, and to explore how it can benefit your organization, connect with us today. Let’s discuss how you can leverage AI to drive efficiency, innovation, and success in your servicing and asset management operations. 

For a more comprehensive analysis of Gen AI in Lending and Servicing,
request a copy of our brochure here.