The Power of AI: Reshaping CRE One Agent at a Time
CPE Voices panelists share strategies and best practices for using the technology in a wide-ranging conversation.
As AI continues to shape the commercial real estate industry, companies are finding new ways to implement the technology. From automating workflows to enhancing data analysis and streamlining day-to-day processes, AI is already gaining traction across the sector.
During Commercial Property Executive’s Harnessing the Power of AI in CRE webinar, panelists came together to discuss where the technology has been beneficial, how it has reshaped the industry and things to look out for when using AI-powered tools. In a discussion moderated by Editorial Director Suzann Silverman, they agreed that while AI adoption is already underway, organizations should start with small, practical wins.
“We are beyond the hype stage in terms of technology,” Kapil Lahoti, chief digital & technology officer at CBRE, said during the webinar. “AI is already impacting our lives daily, so we should get over that and just focus more of how we leverage it.”
Throughout the discussion, the panelists pointed to areas where AI can be implemented, how to implement it, balance what the technology can do and where human interaction is still needed.
Starting small and scalable
When first integrating AI into business processes, Lahoti signaled that the most common successes are where outcomes are already understood. A specific example he used was document-heavy workflows and how AI is able to get through the process faster and extract details from the documents to reduce friction.
Cliff Taylor, director at Yardi, believes the same. During the webinar he pointed out that mundane tasks such as transcribing meeting notes or closing work orders are a nice starting point for AI use, before leaning into the snowball effect.
“Find that early win and then let that roll into the next one and the next one,” Taylor said. “It goes to changing the hearts and the minds of your team and how they work with AI.”
Putting guardrails in place
However, using AI is not a catch-all solution, as there is still the chance for hallucinations, bias and other mistakes.
Tim Savage, clinical assistant professor at NYU Schack Institute of Real Estate, emphasized the importance of understanding the strengths and limitations of algorithmic decision-making. He added that clean data remains essential, as AI tools can amplify errors just as easily as efficiencies when underlying information is flawed.
One of the ways to address this goes back to Lahoti’s point of starting in areas where you know the outcomes. By starting there, companies can more easily test performance and reduce the risk of incorrect results. Separately, another guardrail that Lahoti recommends is governance and ensuring you are compliant and using data correctly.
READ ALSO: How Will AI Reshape CRE Financing?
When deciding to use a large language model product, Lahoti says not to be in a rush and select the first option available. With a technology like this it makes sense to have a person setting up the model and the context while also checking the output.
Giving the AI agent feedback and correcting it, while being strategic about where to apply this, is also a way to think about the process. With these checks in place, hallucinations with AI can be reduced.
“Having a human in the loop is really important,” Taylor said. “We can test, we can challenge and we can revert many times. That way the outputs are much more reliable.”
Measuring impact and redefining roles
Starting with a smaller scale of implementation leads teams to measure their return on investment in a different way. By removing end-to-end friction, as Lahoti pointed out, there is a material ROI as AI is not used in a one-off use case. He also noted that this technology creates more flexibility, which is beneficial for people working in real estate.
One specific example Taylor used to showcase ROI was looking at how much time is saved. If AI could streamline the way property managers look at documents or press ‘approve’ on an invoice, this can save workers time on mundane, repetitive tasks.
“It can be as simple as the time-to-knowledge is sped up,” he said. “That high-volume, low intelligence work is impactful for a business, because you no longer have holdups.”
As these more routine tasks are able to be handed off to technology, this allows individuals in the commercial real estate industry to become more specialized in other areas. Across the panel, it was discussed how the industry roles are changing.
Savage said that domain knowledge is preeminent in this industry and that will continue to be the case especially since, as he put it, “this is an industry that is relationship-based.”
Taylor mentioned that critical thinking skills are becoming more essential among roles, since AI outputs can be wrong. Individuals need the skills to challenge this.
Lahoti said that the repetitive jobs will change as fewer people are needed to do the work, and roles will continue to shift.
“I can’t wait until it takes over some of the jobs that I do,” Taylor said. “I think that evolution is positive, it’s uncomfortable, but it is positive.”
You must be logged in to post a comment.