AI Sentiment in CRE – ICSC Review – Executive Report

1. Executive Summary

This report provides a comprehensive overview of the key observations and sentiments related to Artificial Intelligence (AI) in the Commercial Real Estate (CRE) industry gathered at ICSC Las Vegas 2025 [Insert ICSC Las Vegas 2025 link]. Over the course of the conference, hundreds of industry professionals visited our AI-focused booth, offering a candid look at where the industry stands on AI adoption. Key findings include:

• Cautious Optimism: The general mood was one of cautious optimism. Many CRE professionals are intrigued by AI’s potential to streamline operations and uncover insights, yet broad adoption is in early stages. Surveys indicate over 72% of real estate owners/investors globally are committing or planning to commit funds to AI solutions, but more than half of CRE firms are still only evaluating AI rather than deploying it. This underscores a growing interest tempered by prudence.

• Individual vs. Enterprise Adoption: We observed a gap between individual experimentation and enterprise-wide implementation. On the individual level, many professionals have begun using AI tools (like ChatGPT) informally to assist with tasks. At the enterprise level, however, organizational strategy lags – only ~23% of CRE companies are actively using AI in operations, while 55% remain in exploratory phases. This reflects a bottom-up curiosity meeting top-down caution.

• Prevailing Concerns: Attendees consistently voiced concerns around security, accuracy, and integration of AI. Data privacy risks and the possibility of AI “hallucinations” (incorrect outputs) were frequently mentioned barriers to trust. In fact, industry research shows 65% of CRE professionals worry about data security and 68% about unintended errors with AI. Many are also unsure how AI tools would integrate with legacy property management systems, and some face internal resistance from stakeholders hesitant about change.

• Awareness Gaps: There is a need for education on advanced AI concepts. While most attendees had heard of general AI or chatbots, far fewer understood AI agents or more autonomous AI systems. Over half of CRE professionals admit they know little to nothing about AI’s workings, highlighting an industry knowledge gap that must be addressed for AI initiatives to succeed.

• Emerging Leaders: A subset of forward-thinking CRE firms and executives are moving ahead aggressively with AI. These innovators – including several large brokerages and tech-savvy mid-sized operators – are running pilot projects, investing in proprietary AI models, and even hiring dedicated AI talent to spearhead transformation. Their experiences serve as valuable case studies for the broader industry.

• Use Cases and Opportunities: Conversations at the booth revealed a range of practical AI use cases gaining traction. These include lease abstraction, AI-assisted communications, market analytics, automated payment processing, and enhancing tenant experiences through AI chatbots and voice assistants. Notably, tasks like lease data extraction are seeing 70–90% reductions in processing time with AI assistance, pointing to significant efficiency gains. Each use case carries its own benefits and considerations, detailed later in this report.

• Strategic Recommendations: Finally, we distill lessons learned and offer strategic recommendations for CRE executives. High-level guidance includes starting with low-risk pilot projects in proven use cases, investing in data quality and IT infrastructure for AI, addressing security and compliance up front, fostering internal AI champions, and developing a clear AI roadmap and governance policy (given only ~30% of firms have formal AI policies today). By taking a proactive yet measured approach, CRE organizations can harness AI’s advantages while mitigating its risks.

Overall, AI was top-of-mind at ICSC 2025 – viewed as a transformative force that will increasingly shape how deals are made, properties are managed, and companies compete. The consensus among industry leaders is that ignoring AI is not an option; the challenge is determining how best to embrace it. The sections below provide a deeper dive into the sentiments shared by attendees, the challenges and success stories discussed, and actionable insights for executives charting their AI strategy.

2. General Sentiment on AI in CRE

At the conference, the general sentiment toward AI in commercial real estate was a blend of enthusiasm for its promise and caution about its pitfalls. Attendees widely agreed that AI has the potential to transform key aspects of CRE – from speeding up deal analysis to automating routine workflows – yet many firms are treading carefully. Two contrasting themes emerged repeatedly in our conversations:

• Individual vs. Enterprise-Level Adoption: There is a noticeable gap between personal and organizational adoption of AI. On an individual level, many CRE professionals are experimenting with AI tools in their day-to-day work. For example, analysts and brokers mentioned using ChatGPT or similar tools to draft emails, summarize market data, or generate report outlines. These individual users often take initiative on their own, without a formal mandate, to save time on tasks like writing marketing copy or reviewing leases. However, at the enterprise level, most organizations have not yet fully embraced AI. Company-wide AI deployments are still rare – several attendees noted that their firms have “AI committees” or are running limited pilots, but broad rollout is pending more proof of value and security. In fact, more than half of CRE companies are only in the evaluation stage with AI and have not implemented it at scale. This indicates a scenario where bottom-up innovation is outpacing top-down strategy. The interest of individual employees is high, but many companies have yet to catch up with official tools, training, or budget allocations for AI. One attendee summed it up as, [Insert quote from CRE executive here], highlighting how personal curiosity is driving experimentation even as corporate policy lags.

• Caution vs. Experimentation: We observed an interesting split between those who urge caution and those pushing for experimentation. A significant subset of executives expressed a cautious stance – they are intrigued by AI’s potential but wary of rushing in. Their caution stems from concerns over reliability, data security, and the unpredictability of new AI tools. Many of these leaders prefer to wait for clearer regulatory guidance and for AI solutions to mature. In contrast, a number of forward-thinking professionals (often in innovation or tech roles) advocated for active experimentation. They argue that the best way to understand AI’s value is to pilot it in real scenarios, learn from small failures, and iterate. These proponents of experimentation pointed out that competitors and other industries are moving quickly, and waiting too long could mean falling behind. The overall mood could thus be described as “optimistically cautious.” Most attendees believe AI will significantly impact CRE in the next few years – a view echoed by broader surveys where 92% of C-suite leaders expect AI to change how the workforce operates within five years. Yet, the immediate approach is often incremental: try a few use cases, build internal knowledge, and expand from there if results are positive. In short, everyone is interested in AI, but the industry is calibrating the pace of adoption to balance opportunity with risk.

3. Awareness and Understanding of AI Agents

Despite AI being a buzzword, we found that awareness and understanding of advanced AI concepts vary widely among CRE professionals. In particular, the notion of AI “agents” – autonomous AI systems that can perform tasks or make decisions on a user’s behalf – was not widely understood by the average attendee.

Most people we spoke with were familiar with basic AI applications like chatbots or predictive analytics. Nearly everyone had heard of ChatGPT or similar large language models, and many had tried these tools for simple tasks. However, when conversations dug deeper into concepts such as agentic AI (AI systems that proactively take actions, string together multiple steps, or simulate human-like decision-making processes), attendees often needed additional context and explanation.

Several observations stood out:

• “AI Agents” vs. Chatbots: Many equated AI primarily with chatbots or question-answering tools. The idea that AI agents could autonomously handle multi-step processes (for example, scanning incoming tenant emails, determining priority, drafting responses, and scheduling follow-ups without human intervention) was novel to a lot of attendees. When we demonstrated examples of AI agents orchestrating tasks end-to-end, the response was a mix of fascination and surprise. It became clear that while the term “AI agent” might be common in tech circles, in CRE it remains a nascent concept.

• Knowledge Gaps: There is a notable gap in AI literacy. According to a recent industry survey, 52% of CRE professionals admit they know very little or nothing about AI technology. Our on-site interactions reflected this statistic – over half of the booth visitors needed clarification on how AI algorithms learn, what distinguishes one AI tool from another, and what limitations (bias, data needs, etc.) exist. Only a minority described themselves as well-versed in AI, often those with a tech background or those whose companies had provided AI training. We frequently found ourselves clarifying jargon and foundational concepts (e.g., explaining terms like “machine learning,” “natural language processing,” or the concept of AI hallucinations).

• Curiosity to Learn: On a positive note, the curiosity level was high. Once introduced to the concept of AI agents, many attendees wanted to know more. Questions like “How exactly would an AI agent work in a leasing context?” or “What’s required to set up an AI agent for our internal processes?” were common. This suggests that while awareness is limited, the appetite for understanding and learning is strong. People were especially interested in real examples and case studies of AI agents in action (if any exist yet in CRE), indicating that providing education and demos can rapidly increase understanding.

• Role of Vendors and Demonstrations: Because in-house knowledge is limited, attendees seemed to rely on vendors, demos, and peers to learn about AI capabilities. Our booth’s demonstrations of AI-driven lease abstraction and automated customer service were eye-openers for many. Some remarked that seeing AI work on real estate tasks “made it more tangible” than abstract talk. [Insert quote from CRE executive here] could be used as a placeholder for a direct comment from an intrigued attendee. The key takeaway is that hands-on exposure significantly improves comprehension of AI’s practical uses.

• Implication for AI Adoption: This varying level of awareness means that any enterprise rolling out AI solutions needs to include a strong educational component. Companies cannot assume that their workforce (or even their management) fully grasp what AI can and cannot do. Several attendees from firms that had started AI pilots noted internal teach-ins and workshops as critical to get everyone on the same page. Without building this understanding, attempts to implement AI might face confusion or misuse. In summary, bridging the knowledge gap is an essential step – the industry must invest in training and clear communication about AI tools (including what “AI agents” are) to ensure successful adoption.

4. Concerns Voiced by Attendees

While optimism about AI’s possibilities was evident, attendees were equally vocal about their concerns. Across dozens of conversations, a set of common worries emerged, reflecting the practical challenges CRE professionals foresee in adopting AI. The most frequently cited concerns were:

• Security and Trust: Data security was the number one concern for many executives and IT professionals at the conference. CRE firms handle sensitive information – tenant data, contracts, financial details – and the idea of feeding this into an AI (especially a third-party or cloud-based AI) raised red flags. Attendees questioned how to safeguard proprietary data when using AI platforms and were wary of potential breaches or misuse. Trust in AI outputs also falls under this umbrella. People asked, “Can I trust the AI’s answer enough to act on it for an investment decision or a legal document?” Earning that trust is a hurdle. This aligns with industry findings that 65% of CRE professionals worry about privacy and security risks with AI. Some company representatives noted they have strict policies banning employees from inputting confidential data into public AI tools – underscoring a widespread caution. Transparency is another facet of trust: if an AI recommends rejecting a tenant or altering a portfolio, stakeholders want to know why. Without clear explanations, the “black box” nature of AI makes some uneasy.

• Hallucinations and Accuracy: The issue of AI accuracy – and conversely, the risk of AI “hallucinations” (confidently providing incorrect information) – was brought up repeatedly. CRE professionals are keenly aware that even small errors in things like lease abstracts, financial models, or market research could have serious repercussions. Attendees recounted instances (often second-hand stories) of tools like ChatGPT producing believable yet wrong outputs. This makes them hesitant to rely on AI for mission-critical tasks without human verification. One asset manager quipped that an AI hallucinating is like “a rookie analyst making things up” – unacceptable if unchecked. Indeed, unpredictability or unintended errors were cited by ~68% as a top concern in a 2024 survey. To mitigate this, many suggested that AI outputs must be rigorously reviewed by staff. The prevailing attitude is that AI can draft or analyze, but a human needs to validate. Until accuracy can be guaranteed or errors meaningfully reduced, this concern will remain a barrier to full trust in AI-driven decisions.

• Integration with Existing Tech Stacks: CRE companies have substantial existing IT investments – property management systems (like Yardi, MRI), CRMs, accounting software, data warehouses, etc. A significant concern is how AI solutions will integrate with these legacy systems. Attendees questioned whether new AI tools would “play nice” with their current stack or create data silos and workflow fragmentation. Many do not want a standalone AI app that doesn’t connect to what they already use. Integration challenges are both technical (APIs, data formats) and organizational (ensuring consistency across platforms). Notably, over 60% of real estate firms acknowledge they still rely on legacy technology and find it difficult to adopt emerging tech like AI. This echoes what we heard: the excitement about AI is often dampened by the practical reality of older systems and databases that weren’t designed for AI. Some IT leaders at the booth emphasized the need for middleware or platforms that can bridge AI with their core databases securely. Without smooth integration, they fear AI projects could become cumbersome one-off solutions that don’t scale.

• Internal Organizational Resistance: Beyond technical issues, cultural and organizational resistance is a serious concern. Change is hard in any industry, and CRE – which has long been relationship-driven and often traditional in its workflows – is no exception. Attendees spoke of colleagues (and occasionally senior leaders) who are skeptical of AI’s benefits or fearful of its implications. Part of this is fear of job displacement; although only about 25% of CRE professionals list job loss due to AI as a primary worry, it still exists and can foster quiet pushback. For example, property management teams might resist an AI tool that automates tenant communications, perceiving it as encroaching on their role. Another aspect is simply inertia and comfort with “the way we’ve always done it.” One leasing director shared that convincing veteran team members to even try an AI assistant was an uphill battle. Additionally, lack of clear direction from leadership can breed resistance; if executives aren’t visibly championing the effort, middle managers may not prioritize it. In some cases, we heard that legal and compliance departments themselves tap the brakes, concerned about regulatory risks or liability with AI. This internal friction means that even if the technology is ready, human factors can slow down adoption. The consensus among progressive attendees is that change management and clear communication are as important as the technology itself in any AI rollout.

These concerns underline that while AI technology is advancing rapidly, winning hearts and minds within organizations is the real challenge. Any AI initiative in CRE must proactively address security protocols, establish accuracy benchmarks and oversight, ensure technical integration paths, and have a strategy for organizational change (including training and policy). The next sections discuss how some leaders are navigating these challenges and which areas of AI application drew the most interest as promising starting points.

5. Emerging Leaders and Forward Thinkers

Even as many firms take a cautious approach, a number of emerging leaders in the CRE industry are distinguishing themselves by embracing AI early and boldly. At ICSC 2025, we had the opportunity to interact with and hear about those on the forefront of AI adoption in real estate. These forward thinkers – whether entire companies or individual change agents within organizations – provide a glimpse into CRE’s tech-driven future. Key characteristics and examples of these pioneers include:

• Active Pilot Programs: Forward-thinking companies are not just talking about AI; they’re actively experimenting. Several large firms (including some top global brokerages and institutional owners) have launched pilot programs across various functions. For instance, we heard of a retail REIT that is piloting an AI-driven analytics tool for site selection, and a brokerage firm that built an internal chatbot to answer brokers’ questions by pulling from proprietary research reports. These pilots often start small (limited data sets or one department) but are designed to scale up if successful. The leaders behind them shared that the lessons from hands-on pilots are invaluable for understanding AI’s real capabilities and challenges.

• Hiring and Investment in Expertise: A distinguishing factor of AI leaders is their willingness to invest in expertise. Some firms have hired data scientists, ML engineers, or designated “Head of AI/Innovation” roles to spearhead their initiatives. According to industry insight, a few real estate companies are already bringing in specialized talent to transform workflows. One forward-thinking CEO we spoke with mentioned forming a small AI task force internally, composed of tech-savvy team members from IT, operations, and business units, to drive AI projects and vendor selection. Additionally, partnerships with PropTech startups are common – innovative firms are investing in or closely collaborating with AI startups to co-develop solutions tailored to CRE needs. This gives them early access to new technology and a say in product development.

• Strategic Vision: Emerging leaders approach AI with a strategic mindset. Rather than ad-hoc tool adoption, they have a roadmap of where they believe AI can add the most value in their business model. At the conference, these visionaries often spoke about identifying core business problems (e.g., the time spent on manual data entry, or the inconsistency of service in property management) and then seeking AI solutions for those specific issues. One could sense an urgency in their perspective – a belief that AI is a competitive differentiator. As one innovation director put it, [Insert quote from CRE executive here], underscoring that they see AI not just as tech hype, but as a means to redefine how they operate and serve clients. This clarity of purpose sets them apart from firms still in the AI brainstorming phase.

• Early Wins and Knowledge Sharing: Many forward thinkers are already reporting early wins. In our discussions, some shared measurable results: for example, a regional investment firm automated parts of its underwriting process using an AI model and saw deal analysis time drop from weeks to days. Another attendee from a leading brokerage noted that their internal AI tool for research saved hundreds of analyst hours by quickly summarizing market data. These success stories, even if preliminary, are building momentum and internal buy-in at their organizations. Importantly, these leaders are also keen on knowledge sharing. We noticed a community forming – people swapped notes on which vendors worked well, what pitfalls to avoid, and how to set up governance. At panels and informal meetups during ICSC, the companies ahead of the curve often openly discussed their journeys, helping others learn. This “rising tide lifts all boats” mentality is accelerating learning across the industry. It’s likely that the practices and standards set by these early adopters (in areas like ethical AI use, data management, etc.) will inform broader industry guidelines in the near future.

• Culture of Innovation: Finally, emerging AI leaders all have one thing in common: a culture that encourages innovation and calculated risk. Leadership at these firms tends to be tech-positive and future-oriented. They celebrate experimentation and are not afraid of small failures in pursuit of long-term gains. This culture empowers employees to propose and pilot new ideas without fear. For example, we learned about a mid-sized retail property owner whose executive team allocated a special budget for employees to spend on trying new tech each quarter – essentially seeding grassroots innovation. This kind of environment is fertile ground for AI adoption, as it allows organic development of use cases from within the business.

In summary, the forward thinkers in CRE are already mapping the path that others will likely follow. They exemplify how committing resources (both human and financial), setting clear objectives, and fostering an innovative culture can yield tangible benefits from AI. Their experiences suggest that while challenges exist (as discussed in the concerns section), they can be managed with the right approach. For executives observing from the sidelines, these leaders provide proof that AI in CRE is not just theoretical – it’s happening now, and it’s delivering value for those bold enough to pursue it.

6. Small and Mid-Sized CRE Players: Challenges and Opportunities

The conversations at our booth included many representatives from small and mid-sized CRE firms, and their perspective on AI is uniquely shaped by their scale. These players often lack the resources of large corporations, which presents both challenges and certain advantages when it comes to AI adoption:

• Challenges: For smaller CRE companies, resource constraints are the most immediate hurdle. Limited budgets make it difficult to invest in cutting-edge AI platforms or hire specialized data talent. Unlike a global REIT or brokerage, a mid-sized firm might not have a dedicated IT innovation team; in some cases, one person (or an external consultant) handles all tech matters. This raises concerns about how to even start with AI – who will lead the effort, and how to afford it? Additionally, access to data can be a challenge. Larger firms sit on vast troves of historical leases, market research, and portfolio data that can fuel AI models, whereas smaller firms may not have big datasets readily available (or their data is scattered in spreadsheets and PDFs, needing cleanup). Another challenge voiced is competing priorities – at a small firm, people wear multiple hats and may not have the bandwidth to experiment with AI tools that are not mission-critical yet. There’s also a degree of skepticism or risk-aversion observed: a failed tech project can sting more for a lean operation than for a big company, so the margin for error is thin. As one attendee from a boutique brokerage noted, convincing ownership to fund an AI pilot means showing a very clear ROI, or it won’t get green-lit. Finally, small/mid firms worry about keeping up with the competition – there’s a fear that bigger competitors with more resources might pull ahead in AI, but also a fear of investing in the wrong solution and wasting precious funds.

• Opportunities: On the flip side, smaller players also have unique opportunities in adopting AI. Several mid-sized firm representatives pointed out that being smaller can mean more agility and less bureaucracy. Decisions can be made and implemented faster when fewer stakeholders are involved. For instance, a local property management company might not need multiple rounds of approval to try an AI-based maintenance request chatbot – if the owner or managing director is tech-forward, they can pilot it in a matter of weeks. This nimbleness allows smaller firms to experiment with off-the-shelf AI tools quickly and relatively quietly, learning what works without the high stakes of a massive rollout. Moreover, because they often don’t have extensive legacy systems, some small firms can leapfrog straight to modern, cloud-based solutions. One attendee shared how their mid-sized investment firm skipped on-premise data servers entirely, using a cloud CRM and now integrating an AI add-on to automatically log and analyze deal notes – something that might be harder to do in a heavily entrenched legacy IT environment. Niche focus is another opportunity: smaller CRE companies sometimes specialize in a particular segment or market (e.g., self-storage facilities in one region). This focus means they can look for an AI solution precisely tailored to their niche, potentially giving them an outsized advantage there. For example, a mid-size retail leasing company could adopt an AI tool specifically for optimizing tenant mix in community shopping centers, a very specific use case that larger generalist firms might overlook. Lastly, close-knit teams in smaller firms can make tech adoption smoother – training 20 people on a new AI tool is often easier than training 2,000. There’s a chance to build an internal culture of innovation from the ground up, as there are fewer layers to penetrate with new ideas.

• Collaboration and External Help: Many small and mid-size attendees were interested in partnerships and third-party solutions. Knowing their limitations, they see value in leaning on vendors, industry associations, or even consortiums. For example, some mentioned joining pilot programs offered by PropTech vendors that cater to mid-market firms with sandbox trials or discounted implementations. Others talked about pooling resources – informally sharing learnings with peer firms or through local CRE tech meetups. This collaborative approach can help spread out costs and knowledge. In fact, one theme was that AI could be an equalizer if leveraged properly: by using readily available AI tools (like cloud AI services or even free GPT-based applications), a small firm can automate tasks just like a big firm can. The key is choosing the right tools and focusing on the highest-impact areas first.

• Outlook: Overall, small and mid-sized CRE firms are in an interesting position. They face headwinds in terms of money and manpower for AI, but they aren’t burdened by large bureaucracies or legacy thinking to the same extent. Those we spoke with who had dabbled in AI reported even modest wins (like automating part of their marketing outreach or using AI to scan property expenses for anomalies) were getting noticed and encouraging them to do more. The sentiment was that doing nothing is not a safe option – even mid-market players feel the pressure to modernize. But the approach is pragmatic: start small, use what’s available, and focus on tangible improvements. If they can capitalize on their agility and tap external support, smaller CRE firms could find that AI helps them punch above their weight in efficiency and client service, closing some of the gap with larger competitors.

7. Use Cases Highlighted by Attendees

Perhaps the most illuminating insights from ICSC Las Vegas 2025 came from the specific AI use cases that attendees were interested in or actively exploring. Through hundreds of booth conversations, certain applications of AI in CRE repeatedly came up as promising or high-value. Below is a summary of the top use cases highlighted by industry professionals, along with context on how each is being considered:

• Lease Abstraction: Extracting key information from leases and legal documents was one of the most frequently cited use cases for AI. Attendees, especially those in asset management and brokerage, expressed excitement about using AI to automatically read leases (often long and complex) and distill the important terms – rent schedules, expirations, options, clauses, etc. The manual process of lease abstraction is labor-intensive, often taking several hours per lease for a human reviewer. AI tools now offer the ability to perform this in minutes. At the conference, some visitors shared that they were piloting or considering solutions that can scan PDF leases and output a structured summary. The benefits are clear: faster due diligence, up-to-date lease databases, and reduced human error. We heard anecdotes like “our team could only get through a handful of leases a day before; with AI we’re aiming to review dozens per day.” In fact, early adopters have reported 70–90% reductions in time spent per lease by using AI – for example, going from a 3-hour manual read to a few minutes of AI processing plus quick validation. However, attendees also noted the need to verify AI-extracted data for accuracy (tying back to the hallucination concern). Still, even with verification, the consensus was that AI-powered lease abstraction is a near-term “quick win” that can free professionals from tedious document review and allow them to focus on higher-value analysis and decision-making.

• AI-Powered Email and Communications: Another prevalent use case revolves around automating and enhancing communications, particularly email. CRE professionals deal with a high volume of emails daily – inquiries from tenants, updates from property managers, negotiations with brokers, etc. Attendees showed interest in AI tools that can draft responses, prioritize messages, and even analyze sentiment. For instance, an asset manager might use an AI email assistant to draft a response to a tenant’s question about common area maintenance charges, ensuring the tone is professional and the content accurate (pulling in lease data as needed). Brokers discussed using AI to help write outreach emails to potential tenants or investors, tailoring the content based on data (like the recipient’s profile or market trends). Some property management teams are experimenting with chatbot-like agents integrated with email or text messaging to handle routine tenant communications – e.g., an AI that can respond 24/7 to simple questions like “What’s the procedure for a maintenance request?” or send automated reminders for rent due dates. The goal is to improve responsiveness and consistency of communication while reducing the manual load on staff. Attendees did caution that any AI-generated communication must be reviewed to maintain quality and accuracy, especially in a client-service business where tone and clarity matter. But overall, many saw AI as a valuable “assistant” for the myriad micro-communications that happen in CRE, from crafting polished property descriptions to translating legal jargon for a client email.

• Community-Level Market Intelligence: CRE professionals often need to keep a finger on the pulse of local markets and communities – knowing the latest developments, demographics shifts, or even sentiment in the area around their properties. A forward-looking use case raised at the conference was leveraging AI to gather and synthesize hyper-local market intelligence. This might include using AI to scrape local news, social media, and public data for signals about a neighborhood (for example, detecting an uptick in foot traffic from geo-tagged social posts, or identifying if a new competitor business is generating buzz). Attendees imagined AI agents that could continuously monitor a trade area and alert them to relevant changes – such as emerging retail trends, changes in consumer sentiment, or community concerns that might affect a shopping center’s performance. For instance, an AI could analyze city council meeting notes and flag proposed zoning changes near one’s property, or analyze mobility data to see how a new development is impacting traffic patterns. Some larger firms already have GIS and data teams; AI could supercharge these efforts by automating the data collection and initial analysis. One retail strategist we spoke with said they’d love an AI tool that provides a “daily brief” on each of their key markets, compiled from dozens of sources that would be impossible to manually track. This use case is still emerging, and few had fully implemented it, but interest was high. The challenge, as noted in discussions, is ensuring the AI filters signal from noise and provides truly actionable insights (not just data overload). Nonetheless, the sentiment was that community-level intelligence is a differentiator in leasing and development decisions, and AI could offer a competitive edge by uncovering local insights faster and more comprehensively than traditional methods.

• Automated Payments and Backend Ops: Back-office operations in CRE – such as accounts payable/receivable, invoice processing, and financial reconciliation – are ripe for AI-driven automation, according to many attendees. Property management firms in particular dealt with large volumes of invoices (from vendors, contractors, utility bills) and payments (rents from tenants, CAM recoveries, etc.). The idea of using AI (often in tandem with robotic process automation, RPA) to handle these repetitive tasks generated a lot of interest. For example, an AI system could read incoming invoices (using OCR to capture details), code them to the correct expense categories, flag any anomalies (like a charge that’s much higher than usual), and even initiate payment or approval workflows – all with minimal human intervention. Similarly, on the receivables side, AI can match rent payments to the correct leases, send automated receipts, or nudge tenants with friendly reminders if a payment is missed. Attendees who manage portfolio finances expressed that reducing human error and improving efficiency in these processes would save time and money. One controller mentioned that their team still spends hours every month cross-checking spreadsheets for reconciliations, a task they feel an AI could do in minutes accurately. Additionally, AI can help with financial forecasting and variance analysis by learning from historical data – for instance, predicting which tenants might pay late or which expenses tend to spike seasonally. Some concerns included ensuring financial data stays secure and that AI algorithms are transparent enough for audit requirements. However, overall, the prospect of automating backend operations was appealing as a way to free staff from mundane tasks and allow them to focus on more strategic finance work (like analysis and budgeting). Attendees cited this as a clear area where ROI could be measured in time saved and fewer errors.

• Proprietary LLMs (Custom AI Models): As AI becomes more ingrained, some of the larger and more tech-savvy firms are looking into developing proprietary large language models (LLMs) or custom AI trained on their unique data. This use case was especially noted by attendees from organizations with substantial internal data repositories – for example, decades of market research reports, lease abstracts, tenant surveys, legal contracts, and email archives. The vision is to have an internal AI system, essentially a company-specific “ChatGPT”, that employees can query for insights or documents. For instance, instead of digging through files, a leasing agent could ask this AI, “What was the average rent for 5-year retail leases in our Midwest portfolio signed in the last 12 months?” and get an instant answer sourced from internal data. Or a lawyer could ask, “Find clauses related to force majeure in all our office leases,” and quickly get the relevant excerpts. By training an LLM on proprietary data, firms keep the knowledge in-house and potentially avoid some security issues (since the model can be kept private). Attendees discussed the significant efficiency and knowledge management benefits: institutional knowledge that currently resides in siloed documents or in veterans’ heads could be made accessible to anyone via AI query. However, building a proprietary LLM is non-trivial – it requires technical expertise, computing resources, and careful curation of training data. A few forward-looking firms have begun experimenting with this by using third-party platforms that allow fine-tuning AI on one’s data. The consensus at the booth was that while not every company will train its own model from scratch, many are interested in customizing AI to their domain. Starting perhaps with smaller “domain-specific” models (like an AI only for leases, or only for market data) and then potentially expanding. This use case reflects a desire for AI tools that understand context specific to CRE companies, as out-of-the-box AI might not grasp the nuances without training (for example, knowing that “CAM” means common area maintenance in this context, etc.). Proprietary LLMs could be the key to that deeper level of understanding and relevance.

• Consolidated Market Intelligence Tools: CRE professionals often juggle multiple data sources – economic data, market vacancy rates, trend reports, competitor analysis, news, etc. A strong theme was the wish for AI to power a consolidated market intelligence platform. Attendees envisioned an AI-driven dashboard or assistant that aggregates data from various subscriptions and public sources and then distills it into actionable insights. For instance, such a tool might pull in data from brokers’ market reports, combine it with foot traffic data and economic indicators, and then generate a summary of “City X Retail Outlook this quarter,” complete with visualizations and key takeaways. Instead of manually reading 10 different PDFs and websites, an executive could rely on the AI tool to give a coherent narrative and highlight anomalies or opportunities. Some even discussed dynamic querying – e.g., ask the platform, “How do this week’s retail sales in our centers compare to the same week last year regionally?” and get an immediate answer pulling from point-of-sale data and historical records. A few PropTech vendors at ICSC showcased early versions of such AI-curated market intelligence, which drew a lot of interest at our booth as well. The appeal is clear: save time and improve decision quality by having a one-stop shop for insights rather than fragmented data hunting. Challenges include ensuring data from different sources can be integrated (which ties back to the integration concern) and verifying that the AI’s synthesis is accurate and not missing context. Nonetheless, many attendees saw this as the future of market research – an AI co-analyst that continuously consumes all relevant information and keeps the team informed in real time.

• AI Voice and Chat for Tenant Experience: Enhancing the tenant experience is a priority for many property owners and managers, and AI is starting to play a role here. Attendees discussed deploying AI in voice assistants and chatbots to interact directly with tenants and visitors. For example, a large shopping center owner mentioned exploring an AI-powered chatbot for their center’s mobile app or website that could answer shopper questions (“What are today’s store hours? Any events this weekend?”) and assist with wayfinding. In an office building context, AI kiosks or voice assistants could help visitors navigate the building or allow tenants to book amenities via simple conversation. From a property management angle, AI chatbots can handle tenant requests or FAQs: a chatbot integrated with the building’s system might let a tenant say, “My office is too cold,” and the AI can log a ticket with maintenance or even adjust smart thermostats if integrated. Voice AI is also being looked at for call centers – instead of pressing numbers in a phone menu, a tenant could explain an issue to an AI that understands and routes it properly. Attendees in property management were cautiously optimistic: they see how this could provide 24/7 support and quick answers, improving satisfaction. However, they stressed that the AI must be robust and accurate – miscommunication by a bot can frustrate tenants. Some have started with limited-scope chatbots (like only for answering basic leasing questions on a website) and plan to expand gradually. Another emerging idea was using AI-driven voice assistants during property tours or in smart buildings – e.g., a prospective tenant can ask the empty office space, via a voice interface, various questions and get immediate data (floor plans, rent terms, etc.). Overall, using AI for tenant-facing interactions is seen as a way to provide fast, consistent service, augmenting the human property management team who can then focus on more complex tenant needs.

• Competitive Tool Mapping Platforms: In the competitive world of CRE, knowing what competitors are doing – where they’re investing, which tenants they’re securing, what tech they’re using – is valuable intelligence. Attendees floated the concept of AI platforms that help map out the competitive landscape more effectively. A couple of interpretations came up: one is location-based competitive mapping, where AI could analyze data to show all competing properties or developments around a target site, and even predict their impact (for example, if a new retail center is opening nearby, what does AI forecast about its draw from your center?). The other interpretation was competitive tech/tools mapping – essentially understanding what tools or strategies competitors are leveraging (like which firm has an AI lease abstraction tool, who’s using drone surveys, etc.). Focusing on the first meaning (location competition), AI can crunch datasets about property locations, tenant rosters, traffic, and demographics to produce a map or report of how saturated a market is and who the main players are. Attendees involved in site selection and acquisitions showed interest in an AI that could, say, take a prospective address and instantly list all relevant nearby properties, their sizes, key tenants, and even an estimate of their performance or foot traffic – tasks that currently involve a lot of manual research using services like CoStar, local records, and news scanning. AI’s ability to cross-validate multiple sources could surface a more complete picture quickly. For the second angle (tech mapping), a few tech-forward attendees noted that AI could assist in aggregating information on what’s being adopted across the industry – essentially competitive benchmarking of innovation. For instance, parsing through press releases, conference presentations, and social media to report that “X% of top retail landlords have implemented some AI chat solution” or identifying which competitor has a partnership with a known PropTech vendor. This can inform a firm’s own strategy (so they’re not left behind). Both use cases are about leveraging AI for faster competitive intel. While still an evolving idea, it was clear that companies want better tools to stay informed in a fast-moving environment, and AI’s speed and analytical power could be a game-changer in competitive awareness.

• FSBO-Style AI-Driven Marketing Tools: “For Sale by Owner” (FSBO) in residential real estate refers to owners selling property without a broker. In CRE, brokerage services are highly valued, but some smaller transactions and owners might consider more DIY approaches if empowered by technology. Attendees discussed AI-driven marketing platforms that could allow owners to handle more of the leasing or sales process themselves, reducing reliance on brokers for certain deals. Imagine an AI tool where a property owner can input details about a vacant space (size, location, ideal tenant type or buyer profile, etc.), and the AI then generates a full marketing package: write-ups, flyers, online listings, even a virtual tour script. It could analyze market data to suggest an asking rent or price, identify a target list of potential tenants/buyers, and maybe even handle outreach via targeted ads or emails. Essentially, a mini-broker in a box. For landlords with small portfolios or unique properties, this could be appealing to save on commissions and take control of the process. Conference attendees saw opportunity here especially for lower-tier assets or subleasing scenarios where hiring a top brokerage team might not be cost-effective. However, they also acknowledged the limitations – relationship management and negotiation are hard to automate, and that’s where brokers excel. Rather than completely replacing brokers, a more plausible scenario discussed was AI handling the marketing prep work and lead generation, then the owner or a broker steps in for the negotiation and closing. Some PropTech startups are indeed working on elements of this (e.g., AI-generated property listings). Attendees were curious but also cautious: the consensus was that AI-driven FSBO tools might empower owners and put pressure on brokers to deliver more value, but they likely won’t eliminate the need for human expertise in complex deals. Still, in the spirit of experimentation, a few landlords said they might try such platforms for simpler transactions. The takeaway is that AI is lowering the barrier for property marketing and could democratize some aspects of deal-making – a trend to watch in the coming years.

• Hands-on Staff Training Solutions: The final use case that got considerable attention was using AI for training and upskilling staff. As new tools and processes emerge (often driven by AI itself), keeping the workforce trained is a big challenge. Attendees were interested in AI solutions that can expedite and enhance training for CRE professionals, whether it’s onboarding new hires or teaching existing employees new systems. One concept is AI-driven interactive training modules: for example, a leasing agent could practice negotiation with an AI that simulates different tenant personalities, providing a safe sandbox to hone skills. Another idea is an AI mentor or coach that employees can ask questions to on the fly. Instead of flipping through manuals, a property manager could query an AI assistant, “How do I handle a tenant’s request to sublease, according to our company policy?” and get an instant, tailored answer. Essentially, AI as a dynamic FAQ and troubleshooting resource that learns from the company’s own policies and past decisions. Some attendees noted their firms are creating internal knowledge bases, and coupling that with an AI Q&A interface would make that knowledge far more accessible day-to-day. Micro-learning through AI was another theme: AI can deliver bite-sized training content (short quizzes, flash cards, or even games) to reinforce learning gradually, adapting to the user’s pace and areas where they need improvement. For CRE roles that require understanding of complex topics (legal clauses, underwriting math, compliance rules), an AI tutor that personalizes the training could significantly shorten learning curves. A practical example shared was using AI to train staff on a new lease management software – the AI can monitor common errors and proactively suggest tips or guide the user through tasks step by step. Attendees seemed enthusiastic about this, as it tackles a real pain point: time and cost of training. However, they also recognized that AI won’t replace the need for human-led training entirely, especially for soft skills or company culture immersion. Instead, AI is seen as a supplement – something that can provide on-demand support and practice, ensuring that formal training “sticks” and employees remain confident as new technologies (like AI itself) are introduced into their workflows.

These use cases illustrate that AI’s reach in CRE is broad – touching virtually every aspect of the business, from front-line leasing and tenant relations to back-office administration and strategic decision-making. It’s worth noting that none of these use cases came across as science fiction; they were either already in trial or well within the realm of possibility today. This groundswell of practical AI applications, as highlighted by attendees, suggests that the industry’s focus is shifting from “What is AI?” to “How can we apply AI to solve this specific problem?”.

The diversity of use cases also implies that different firms will find different entry points for AI depending on their pain points and priorities. For some, automating lease abstracts might be the gateway; for others, an AI chatbot for tenant inquiries could be the first step. The next section synthesizes lessons learned from these conversations and proposes strategic recommendations to help CRE executives navigate this landscape and implement AI in a way that aligns with their business goals.

8. Lessons Learned and Strategic Recommendations for CRE Firms

Drawing together all the insights from ICSC 2025, several key lessons and strategic recommendations emerge for commercial real estate firms looking to capitalize on AI. The path to successful AI adoption involves not just choosing the right technology, but also preparing the organization and aligning projects with business value. Here are the major takeaways and recommendations for CRE executives:

• Start with Clear Use Cases and Quick Wins: A critical lesson is the importance of focus. Rather than attempting a broad, nebulous AI transformation all at once, leading firms pinpoint specific use cases where AI can immediately add value (like the ones highlighted in Section 7). Executives should identify 1-3 high-impact, feasible projects – for example, automating lease abstraction or deploying a pilot chatbot for tenant service – and start there. These projects serve as proofs of concept that can demonstrate ROI and build internal confidence. Quick wins are valuable; even a modest success (e.g., AI saving a few hundred man-hours on a lease review project) creates momentum and buy-in that paves the way for larger initiatives. It’s also advisable to choose use cases aligned with strategic pain points (if data entry is a bottleneck, automate that first; if client responsiveness is an issue, try an AI assistant there). Lesson learned: Avoid doing AI for AI’s sake – tie projects to real business outcomes and knock out some easy wins early.

• Invest in Data Readiness and Integration: AI’s effectiveness is only as good as the data and systems feeding it. One recurring lesson is the need to get your data house in order. Many firms realized during pilots that their data was siloed, unclean, or hard to access – which significantly slowed down or complicated AI deployment. We recommend executives prioritize data consolidation and quality initiatives (e.g., ensuring lease data, property info, and financial records are digitized and centralized). Consider modernizing legacy systems or implementing data lakes that AI tools can plug into. Integration planning is vital: if adopting a new AI tool, involve your IT team early to map out how it will connect with existing software (APIs, data pipelines, etc.). One attendee put it succinctly, [Insert quote from CRE executive here], emphasizing that without good data and integration, AI projects will stumble. Given over 60% of firms cite legacy tech as a challenge in adopting AI, addressing this proactively is a strategic move. Lesson learned: Successful AI adoption often requires parallel investments in IT infrastructure and data management. Don’t neglect the “plumbing” that allows AI to work seamlessly in your environment.

• Address Security and Compliance Up Front: Security concerns can derail AI efforts if not properly managed. A key recommendation is to establish clear policies and safeguards from the outset. If considering a cloud-based AI service, perform due diligence on how it handles data (encryption, data residency, access controls) and involve your legal/compliance teams to ensure regulatory requirements (like GDPR, etc., if applicable) are met. Many companies are creating or updating AI usage policies – guidelines on what data employees can input into public AI tools, for example – to protect confidentiality. It’s noted that as of 2024, only ~30% of CRE firms had AI policies in place, so putting a policy in writing is a recommended first step. Additionally, consider using private or enterprise versions of AI platforms that offer more security, or deploying AI on-premises if the data is extremely sensitive. Another aspect is setting up human oversight where needed: for instance, requiring that any AI-generated client deliverable passes a human review for accuracy and compliance. Lesson learned: By tackling security and trust issues head-on (through policies, tech measures, and oversight), you not only mitigate risk but also build confidence among stakeholders that AI is being used responsibly.

• Foster a Culture of Training and Experimentation: The human element came up time and again – specifically, the need for training and culture change. CRE firms should invest in upskilling their workforce on AI literacy. This doesn’t mean turning everyone into data scientists, but ensuring employees understand the basics of how AI works, its limitations, and how it can assist in their roles. Offering workshops, bringing in experts for seminars, or even simple internal tutorials can make a big difference (remember, more than half of CRE pros currently say they know little about AI). Some firms have created “AI ambassador” teams or pilot groups to champion the technology internally and help peers learn. Encouraging experimentation is also key: leadership should create an environment where trying new tools is welcomed (within agreed guardrails) and where failure in pilot efforts is seen as learning, not a career risk. A tactic here is to allocate a small innovation budget or allow employees a few hours a week to tinker with new solutions. One example from the conference was a company that held an internal AI hackathon for its analysts to come up with creative uses for ChatGPT in their workflow – this not only yielded ideas but also got people comfortable with the technology. Lesson learned: Culture can make or break tech adoption. Executives should actively cultivate an innovative mindset – celebrate those who find new efficiencies with AI, and ensure resources (time, training, encouragement) are given to build internal capability. By demystifying AI through hands-on experience, resistance often fades and enthusiasm grows.

• Lead with a Strategic Vision (Top-Down Support): While bottom-up innovation is valuable, real transformation requires top-down strategic vision. A lesson from companies already ahead is that executive leadership (C-suite or business line heads) needs to clearly articulate why the company is adopting AI and what the ultimate goal looks like. Whether the aim is to become a data-driven organization, to significantly improve customer experience, or to cut operational costs, having those strategic objectives helps align all AI efforts. It also signals to the entire organization that AI isn’t just a pet project in IT – it’s a priority that leadership cares about. Concrete actions at the top might include setting up an AI steering committee, embedding AI goals in the annual business plan, or even tying a portion of performance metrics/bonuses to innovation or efficiency improvements achieved through tech. Moreover, leaders should be visible in their support: for instance, the CEO might mention AI initiatives in company-wide communications or at town halls, reinforcing commitment. A strong leadership voice helps overcome internal resistance; when employees see that leadership is both serious about and supportive of AI (and also realistic about its challenges), they’re more likely to participate constructively. It was noted at the conference that over 90% of C-suite leaders see AI as transformative – but that belief must translate into action and guidance for the organization. Lesson learned: Provide a clear mandate and vision for AI. Align it with the company’s mission (e.g., “We will leverage AI to deliver superior insight and service to our clients”) so that everyone from managers to entry-level staff understands the “north star” they’re working towards with these new tools.

• Partner and Learn Externally: Another recommendation is not to go it alone. The AI field is evolving rapidly, and CRE firms can benefit from external partnerships and knowledge sharing. This could mean partnering with PropTech startups, joining industry consortiums on technology, or working with consulting firms that specialize in AI deployment. For smaller companies, leveraging vendor expertise is especially important since they may not have internal R&D teams. At ICSC, we saw numerous vendors ready to collaborate on pilots or sandbox projects. Engaging with them can accelerate your learning curve (while being mindful to choose partners that align with your needs and values). Additionally, networking with peers – through industry associations or informal meetups – allows executives to swap notes on what’s working or not. Many attendees noted the value of hearing real case studies at the conference. One idea is to create an industry working group among a few non-competing peers to regularly discuss AI experiences. Some mentioned CRE trade organizations possibly facilitating such forums. Lesson learned: AI in CRE is a new frontier for everyone; being plugged into the broader ecosystem can prevent reinventing the wheel. Learn from others’ mistakes and successes, and consider collaborative approaches where possible (even something like co-developing a data standard to make AI integration easier across the industry).

• Be Intentional and Ethical: Finally, a higher-level lesson is to approach AI adoption intentionally and ethically. “Disrupt or be disrupted” was a mantra cited from Deloitte’s perspective – it implies urgency, but it doesn’t mean rushing in carelessly. Having a plan (what processes to transform, what sequence to follow, how to measure success) is vital. Equally important is considering the ethical implications: ensure fairness (the AI doesn’t inadvertently introduce bias in lending or tenant selection, for instance), maintain transparency with stakeholders (if tenants are interacting with a chatbot, should they be informed it’s not a human?), and consider the impact on jobs. Many forward-looking firms are already thinking about how to reskill employees affected by automation so that AI augments rather than purely replaces their roles. This thoughtful approach will help maintain trust both internally (with your team) and externally (with clients and the public). Documenting principles for responsible AI use in your company – such as a commitment to data privacy, accuracy, and human oversight – can provide a compass as you implement new technologies. Lesson learned: An AI strategy is not just technical; it must align with the company’s values and be executed with deliberation. Those who rush implementation without frameworks might face setbacks or backlash, whereas those who are strategic and ethical can set themselves up for sustainable success.

By internalizing these lessons and following these recommendations, CRE firms can significantly improve their odds of a successful AI journey. The overarching theme is balance: balance ambition with caution, innovation with governance, and automation with the human touch. AI has immense potential, but realizing that potential requires more than just buying software – it demands leadership, culture, and strategy. The companies that get this mix right will not only solve immediate problems through AI but will position themselves as industry leaders in the data-driven era of commercial real estate.

9. Conclusion

The discussions and observations from ICSC Las Vegas 2025 made one thing abundantly clear: AI is poised to become a transformative force in commercial real estate, and the industry is both excited and challenged by this prospect. We stand at an inflection point where early experiments are yielding promising results, yet widespread adoption will require careful navigation of concerns and concerted effort in upskilling and strategy.

For CRE executives, the conference’s insights offer a roadmap of what to anticipate. In the short term, we can expect incremental changes – a pilot AI tool here, a process automated there – chipping away at inefficiencies that have long been accepted as status quo in real estate. Lease reviews get faster, customer service a bit more responsive, market research more data-driven. These may start as isolated wins, but collectively they signal a new competitive baseline emerging: speed, intelligence, and adaptability powered by AI.

In the longer term, as comfort with AI grows and success stories multiply, we are likely to see more holistic transformations. Firms that embrace AI strategically will develop smarter portfolios (with insights continuously gleaned by AI), leaner operations (with routine work largely automated), and enhanced stakeholder experiences (tenants, investors, and employees interacting through AI-augmented channels). The gap between AI leaders and laggards in CRE could widen significantly. Those who have invested early in learning and integrating AI may reap disproportionate benefits in efficiency and innovation, while those who remain skeptical or slow to move might find themselves at a disadvantage in everything from winning clients to managing assets optimally.

Yet, the message from ICSC 2025 is not that AI will replace the human element in CRE – rather, it will elevate and support the people in the business. Relationships, creativity, and judgment remain at the heart of real estate. AI tools are enablers, taking away drudgery and providing insights, thereby empowering professionals to focus on high-value activities like strategy, client advisory, and decision-making. As one could gather from multiple attendee conversations, there is a genuine enthusiasm for the idea that AI might finally free teams from the mountains of paperwork and analysis, allowing them to be more proactive and client-centric.

To capture that upside, executives must lead with vision and care. This report emphasized the importance of a professional, strategic approach to AI – one that aligns with corporate goals and addresses the legitimate concerns of employees and clients. Transparency, training, and measured progress are the tools to mitigate fears and build trust in AI systems. Companies that implement AI successfully will likely be those that communicate clearly about its role, set realistic expectations, and maintain a feedback loop to refine how AI is used.

In conclusion, the experience at the AI booth in Las Vegas was both encouraging and enlightening. We saw an industry on the cusp of significant change. The general sentiment might be summed up as: optimism with eyes wide open. There is optimism about AI’s ability to solve real problems in CRE – to make our work faster, our decisions smarter, and our services better. But there is also a sober understanding of the challenges involved – data issues, change management, and trust building, to name a few.

For our organization and others, the strategic imperative is clear: engage with AI proactively and thoughtfully. The recommendations outlined in this report serve as guideposts for that journey. By heeding the lessons from peers and pioneers, and by committing to a culture of continuous learning, CRE executives can navigate the AI revolution with confidence.

The conversations at ICSC 2025 will undoubtedly continue to evolve in the coming years, but one thing is certain: AI is no longer a topic on the fringes of CRE discussions. It’s now center stage, commanding the attention of executives and practitioners alike. Those prepared to embrace this transformation – balancing innovation with wisdom – will lead the industry forward into its next chapter of growth and success.