Why Most Commercial Real Estate Firms Struggle to Deploy AI

Written by:

Rudi Davis

Date:

03 Apr 2026

Share

Share via Facebook

Share via Linkedin

Share via X

Copy to clipboard

Most commercial real estate (CRE) firms are experimenting with artificial intelligence (AI), but few have integrated it into their daily operations. Eglae Recchia, Chief Operating Officer at Keyway, points to a recent survey of about 150 CRE firms. While 45% are piloting AI solutions, only 9% have successfully put them into production workflows. This wide gap highlights the challenge of moving from testing to real-world deployment. It also reveals a core problem: generalized AI tools are not meeting the CRE industry’s specific needs.

Why AI Pilots Stall

Many CRE firms start with popular off-the-shelf tools such as Microsoft Copilot or ChatGPT. These platforms help with broad, generic tasks but fall short of the complex, specialized demands of real estate. As a result, most pilot programs stall before reaching full deployment.

Recchia explains that the gap is not due to skepticism. Firms are actively testing AI but struggle to move beyond experimentation. The core issue is that generalized AI lacks the precision, domain knowledge, and reliability required for CRE processes. These include underwriting, lease abstraction, and portfolio analysis. These are high-stakes tasks where errors can lead to costly mistakes.

“The numbers show excitement about AI, but making the leap from testing to changing processes has been tough,” Recchia says.

General AI Falls Short

Broad AI platforms are rarely designed for the nuances of commercial real estate. These tools can draft emails or summarize documents. They are not, however, equipped to interpret the specialized contracts, financial models, or regulatory requirements that define CRE transactions.

Recchia notes that attempts to use general-purpose AI often encounter problems with accuracy and context. Large language models may generate plausible-sounding responses, but they can also hallucinate, producing information that is incorrect or unsupported by the data. In real estate, where decisions involve millions of dollars and legal risk, even small errors are unacceptable.

“The power of AI in this sector comes from deep expertise,” says Recchia. Keyway, for example, has more than 20 team members focused on technology and data science to ensure their tools reflect real estate’s unique demands.

This recognition is driving the industry away from general AI experiments toward purpose-built, vertical platforms. These specialized tools are trained on CRE-specific data and workflows, reducing the risk of errors and increasing the chances of successful deployment.

“Everyone was trying whatever came along, but now there’s a clear move toward AI tools built specifically for commercial real estate,” Recchia says. Firms are looking for solutions that combine efficiency with the domain knowledge needed to automate real estate tasks accurately.

Accuracy Demands Specialized Tools

Accuracy is a non-negotiable requirement in commercial real estate. Transactions depend on precise financial modeling, market data, and legal documents. Mistakes in underwriting or lease abstraction can result in major financial losses or compliance issues.

General AI tools, which are not calibrated for CRE’s complexity, often generate incomplete or incorrect outputs. This is especially risky in scenarios where a single error could derail a deal or expose a firm to legal liability.

“In commercial real estate, you need concrete answers based on real numbers,” Recchia says. The risk of hallucination or inaccuracy from broad AI tools is too high.

To address this, firms are shifting to vertical AI platforms that integrate domain-specific knowledge and can be validated against established workflows. These solutions are trained on industry documents such as offering memorandums, lease agreements, and loan files, making them more reliable for real estate tasks.

Recchia observes that firms are now prioritizing measurable accuracy and reliability in AI solutions. The industry is moving beyond hype, focusing instead on tools that can prove their value in real-world scenarios.

Vertical AI Closes the Gap

Keyway’s strategy for overcoming the deployment gap is to offer modular AI tools tailored to specific CRE pain points. Rather than pushing a single comprehensive platform, Keyway provides targeted modules for data management, deal execution, and workflow automation. This approach allows clients to start with a single use case, validate the technology, and expand as they see results.

About 30% of Keyway’s clients have already added new modules after their initial deployment, according to Recchia. This incremental approach reflects a broader pattern in the industry: firms are more willing to adopt AI at scale once they have proof that it works in their environment.

The shift toward vertical AI is gaining momentum across the industry. As more firms reach the limits of what general-purpose AI can do, they are seeking out vendors with deep CRE expertise. Successful AI deployment now depends as much on an understanding of real estate workflows and data structures as on technical sophistication.

Barriers to Full Deployment

The fact that only 9% of CRE firms have deployed AI to production indicates the industry is still early in its adoption curve. Firms that manage to bridge the gap stand to gain significant advantages: processing deals faster, reducing manual work, and using resources more efficiently.

However, the low deployment rate also signals persistent barriers. Many firms lack the internal expertise to integrate AI. Others face resistance to changing established processes or struggle to find tools that address their specific needs. As vertical AI platforms become more accessible, more firms are expected to move from pilot to production. This will require a shift in how technology is evaluated and implemented.

Recchia believes the next phase of AI adoption in CRE will be driven by vendors who can demonstrate both technical strength and domain expertise. Firms that continue to rely on general AI platforms may remain stuck in the pilot phase, unable to realize the operational improvements that successful deployments bring.

What Drives AI Adoption

The current deployment gap signals a turning point for AI in commercial real estate. The industry’s initial excitement over general AI tools has given way to a more practical approach, one that demands solutions delivering measurable value in real-world transactions.

As CRE firms become more selective, the market is likely to reward vendors who can combine advanced technology with deep industry knowledge. Those who invest in purpose-built vertical AI platforms will be best positioned to move from experimentation to full-scale adoption as industry expectations for automation, accuracy, and efficiency continue to rise.