An interesting, exciting new generative artificial intelligence (AI) has arrived: ChatGPT. This large language model (LLM) is the chatbot that’s got everyone talking. But what’s all the fuss? Some worry this innovative AI will take people’s jobs. Others think it’s a threat to academic integrity. But still others think this OpenAI chatbot has the potential to revolutionize many industries — in a good way.
So how does it work?
In simple terms, ChatGPT is driven by an LLM — a type of AI capable of drafting memos, presentations, emails, and short essays. If you give it the right prompt, ChatGPT will write you a poem on the topic of your choice, explain quantum mechanics, or recommend a list of marketing best practices.
Huge amounts of data — and computing techniques — power chatbots like ChatGPT or Google’s recently-introduced Bard. These LLMs can mine a nearly limitless quantity of vocabulary and information and understand words in context. This operation helps them imitate speech patterns while impressing users with encyclopedic knowledge.
There are, however, a few caveats:
- The most current ChatGPT version can’t access information more recent than 2021, so if you’re hoping for the latest research or trends, what it generates isn’t necessarily up to date.
- ChatGPT can’t evaluate the quality of the resources it’s using nor does it have the same judgment as a human to identify when something is true or false. So if you’re relying on it to write an email with facts and stats, edit before sending.
- While you can ask ChatGPT to write something in a specific voice or style, emulating a casual tone, for example, or something more businesslike, its word choices are hit or miss — often with amusing (or cringy) results. For example, Vanderbilt used ChatGPT to write an email in reaction to the recent Michigan State University shooting. The message was poorly received, especially after recipients noticed text at the bottom of the email, saying “paraphrase from OpeAI’s ChatGPT AI language model, personal communication, February 15, 2023.”
Potential Use Cases — Including CRE
Used judiciously, however, ChatGPT and other generative AI have multiple use cases across many industries to help with:
- Debugging codes
- Content generation, text summarization, and outline building
- Conducting research
- Nurturing leads
Let’s look at commercial real estate. Some experts think ChatGPT has the power to significantly impact CRE in several areas. For example, AI algorithms can help with property management, including energy or lease management, maintenance scheduling, and rent collection, and facilitate communication between tenants and landlords. When given the appropriate data, ChatGPT can generate analytics and reports and even handle customer service inquiries.
Once AI algorithms are trained to analyze financial reports, market trends, or property listings, CRE pros and investors can use this LLM to make better informed decisions about which properties and markets to explore and invest in. The AI also automates routine, manual tasks including financial reporting so CRE pros can focus on issues and projects that require higher level critical thinking.
More CRE Use-Case Scenarios
AI has other opportunities to shine in CRE, by making the industry more efficient and transparent. Blockchain technology, for example, can help reduce fraud and errors and simplify the process of buying and selling properties by creating a clear, secure ledger of real estate transactions.
CRE pros can use ChatGPT and other AI-powered chatbots to streamline their workflows, including:
- Finding and leasing properties
- Providing prospective clients with more detailed information about available properties
- Offering comparisons among different properties
- Helping with the lease application process
- Assisting with relationship management
Other CRE professionals predict even more significant capabilities in the future, possibly reshaping the brokerage and real estate listing sectors. LLMs already have the potential to improve real estate listing searches. It won’t be long before the AI algorithms will, when given specific parameters, be able to narrow a search based on client preferences to 90% or better accuracy, with brokers reviewing the remaining 10%.
While ChatGPT can only access information online prior to 2021, its creator, OpenAI, recently announced new plugins designed to allow the chatbot to connect to third party applications — including access to real-time information. With these plugins, ChatGPT will gain access to the stock markets and other financial news, sports scores, breaking news and more — just like if you conducted a search on Google or Bing.
Another benefit of this bot? It will help close the knowledge gap between brokers, emphasize the importance of personal relationships and human connections, and more. Will it replace some jobs? Perhaps. But it’s more likely to help — not replace — real estate pros, by freeing them to do what they’re best at — building relationships.
ChatGPT in Action
Jeff Hamann, senior editor of Janover, put ChatGPT to the test in February 2023. This article details real-life examples of the results of three prompts he fed the chatbot. The first prompt asked ChatGPT to list the cost advantages of a client keeping her business in her current area (South Florida) vs relocating to the Philly area. The second prompt offered five different scenarios involving several tenants concerned with issues in the building they rented. The final prompt involved a financing scenario.
The article includes the prompts, ChatGPT’s responses, and Hamann’s analysis. It’s a great illustration of how CRE pros can use this technology to streamline CRE transactions and more.
Are there potential challenges to using an LLM in CRE? Absolutely. This technology comes at a cost to implement and maintain. The AI algorithms aren’t perfect and can make errors, especially when asked to analyze huge quantities of data — it’s important someone review the analyses to fix errors lest they lead to inaccurate investment decisions which could harm investors financially. CRE companies must carefully review, vet and test AI systems before deployment.
However, the future looks bright — and it’s likely that AI and LLMs like ChatGPT have the power to shape the CRE industry, increasing its efficiency, transparency, and customer service delivery.
Are you a commercial real estate investor or looking for a specific property to meet your company’s needs? We invite you to talk to the professionals at CREA United: an organization of CRE professionals from 92 firms representing all disciplines within the CRE industry, from brokers to subcontractors, financial services to security systems, interior designers to architects, movers to IT, and more.