Future AI Trends Impacting Real Estate

I spent the week trying to answer the question: How can I build a real estate development firm with zero human employees? After studying every AI tool in real estate, I found something surprising. Here's what would happen if machines did a developer's job: I can’t stop thinking about how close we are to this reality. From deal sourcing to cost estimation, there's now an AI tool for almost every step. So I designed a hypothetical real estate development company with zero employees: Asimov Partners. The vision: a real estate development firm that builds ground-up multifamily but has zero human employees doing any actual work. For this to work, we’ll use AI to cover: • Deal sourcing • Zoning analysis & test fit • Underwriting • Cost analysis • Buying the site • Raising capital • Permitting and construction • Leasing, management, and sale While the tech isn’t 100% there yet, here’s what I learned: What’s already possible: → AI can analyze thousands of sites simultaneously  → Tools generate floor plans and unit mixes automatically → Financial models build themselves from market data → Cost estimates update in real-time Where we’re stuck: → Lender negotiations still need humans → Construction coordination requires real relationships → Trade management can't be automated → Complex engineering decisions need human oversight The reality: • The most valuable application isn't replacing developers. • It's giving them superpowers to evaluate 100x more opportunities. Here's what this means for development: • Analysts focus on validation, not data entry • Best opportunities go to firms with best algorithms • Automation handles volume, humans handle judgment • Engineers focus on complex decisions, not routine tasks As Suffolk Construction's AI Director told me: “We’re nearly there with automation, but it’s not yet sufficient.” Melek is charged with researching and implementing AI at one of the largest construction contractors in the country, so he sees a lot. The future isn't automated development. It's augmented development. What parts of real estate do you think AI will transform first? Full letter on how I sketched out the build for Asimov Partners here: https://lnkd.in/ed4_gE9k

🤖 𝐀𝐈 𝐢𝐬 𝐄𝐱𝐩𝐚𝐧𝐝𝐢𝐧𝐠—𝐁𝐮𝐭 𝐎𝐟𝐟𝐢𝐜𝐞 𝐒𝐩𝐚𝐜𝐞 𝐍𝐞𝐞𝐝𝐬 𝐀𝐫𝐞 𝐒𝐡𝐫𝐢𝐧𝐤𝐢𝐧𝐠 The AI boom is revolutionizing workflows—and eliminating entire job categories. With fewer junior roles, companies need fewer desks, smaller teams, and significantly less office space. If you wait to adjust your footprint, you risk overpaying for space you no longer need. 🏢 𝐄𝐧𝐭𝐫𝐲-𝐥𝐞𝐯𝐞𝐥 𝐫𝐨𝐥𝐞𝐬 𝐚𝐫𝐞 𝐝𝐢𝐬𝐚𝐩𝐩𝐞𝐚𝐫𝐢𝐧𝐠—𝐚𝐧𝐝 𝐬𝐦𝐚𝐫𝐭 𝐭𝐞𝐧𝐚𝐧𝐭𝐬 𝐚𝐫𝐞 𝐚𝐥𝐫𝐞𝐚𝐝𝐲 𝐬𝐡𝐢𝐟𝐭𝐢𝐧𝐠 𝐬𝐭𝐫𝐚𝐭𝐞𝐠𝐲. Here’s what corporate occupiers need to act on: 🔹 𝐉𝐮𝐧𝐢𝐨𝐫 𝐫𝐨𝐥𝐞𝐬 𝐚𝐫𝐞 𝐯𝐚𝐧𝐢𝐬𝐡𝐢𝐧𝐠: AI is replacing support positions like admins, analysts, and coordinators—shrinking the need for large office footprints. Entire departments are being automated, reducing headcount and spatial requirements. 🔹 𝐂𝐥𝐚𝐬𝐬 𝐁 & 𝐂 𝐛𝐮𝐢𝐥𝐝𝐢𝐧𝐠𝐬 𝐚𝐫𝐞 𝐥𝐨𝐬𝐢𝐧𝐠 𝐫𝐞𝐥𝐞𝐯𝐚𝐧𝐜𝐞: Companies are consolidating into Class A properties that offer amenities, tech infrastructure, and flexibility. Legacy buildings are struggling to attract tenants and may soon become obsolete. 🔹 𝐋𝐚𝐧𝐝𝐥𝐨𝐫𝐝𝐬 𝐚𝐫𝐞 𝐟𝐞𝐞𝐥𝐢𝐧𝐠 𝐭𝐡𝐞 𝐩𝐫𝐞𝐬𝐬𝐮𝐫𝐞: High vacancy rates are forcing landlords to offer better deals, early termination options, and generous concessions. Tenants have unprecedented leverage—but it won’t last forever. 🔹 𝐅𝐥𝐞𝐱𝐢𝐛𝐢𝐥𝐢𝐭𝐲 𝐢𝐬 𝐞𝐬𝐬𝐞𝐧𝐭𝐢𝐚𝐥: In a fast-changing workforce, locking into rigid long-term leases is a liability. Flexible leasing models let companies scale up or down based on real-time needs. 🔹 𝐓𝐡𝐞 𝐨𝐥𝐝 𝐥𝐨𝐠𝐢𝐜 𝐝𝐨𝐞𝐬𝐧’𝐭 𝐚𝐩𝐩𝐥𝐲 𝐚𝐧𝐲𝐦𝐨𝐫𝐞: Planning for intern overflow or buffer space is no longer financially justifiable. Every square foot must contribute to productivity, or it gets cut from the plan. The office isn’t obsolete—but the way we use it has changed for good. #𝐂𝐨𝐫𝐩𝐨𝐫𝐚𝐭𝐞𝐑𝐞𝐚𝐥𝐄𝐬𝐭𝐚𝐭𝐞 #𝐎𝐟𝐟𝐢𝐜𝐞𝐒𝐩𝐚𝐜𝐞 #𝐀𝐈 #𝐂𝐑𝐄 #𝐋𝐞𝐚𝐬𝐞𝐍𝐞𝐠𝐨𝐭𝐢𝐚𝐭𝐢𝐨𝐧

What are non-obvious ways the generative AI boom will impact demand for real estate asset classes? I put out a video earlier this week about how the AI revolution is going to beget a massive real estate industry underneath it, very much like the Internet did, because of an explosion in demand for computational power. In fact, current data center hubs can barely keep up with demand. But what other assets could experience a surge in demand, other than digital infrastructure like data centers and edge computing? AI-ENABLED EFFICIENT BUILDINGS -- Buildings with the proper infrastructure to handle AI will be in demand across asset classes. Much like internet connectivity is table stakes today, such will be the case for AI in the future. On top of improving tenant experience, AI connectivity will also enable buildings to operate more efficiently -- saving on energy bills and avoided fines. RENEWABLE ENERGY -- AI requires a ton of power and cooling. Combined with the increased regulatory scrutiny on real estate's environmental impact and net-zero pledges from giants like Amazon and Google, data centers will need nearby renewable energy and energy storage to power themselves. OFFICE -- Offices with the latest and greatest AI connectivity will be in highest demand from top tenants HEALTHCARE -- Same goes for hospitals and healthcare facilities... LOGISTICS -- Same goes for warehouses and logistics centers... #realestate #artificialintelligence

🤖 AI is coming for CRE underwriting. But not in the way you think. It’s not replacing brokers. It’s not replacing investors. It’s not replacing operators. It’s supercharging them. Here’s what’s already happening 👇 ✅ Underwriting in Minutes, Not Days. AI-powered platforms can now analyze rent rolls, operating statements, and market comps in seconds—spitting out cash flow models and valuations that used to take an analyst 8+ hours. ✅ Better Risk Assessment - No more relying solely on gut instinct or past deals. AI is identifying hidden risks by analyzing: ▪️ Lease default probabilities ▪️ Tenant credit health ▪️ Submarket trends down to the block ▪️ Supply pipelines & absorption rates ✅ Real-Time Deal Screening - What if you could screen 50 deals before lunch—and only spend time on the 3 that truly fit your criteria? That’s not the future. That’s now. ✅ Operational Optimization - AI isn’t stopping at acquisitions. It’s reshaping property operations: ▪️ Predictive maintenance (fix it before it breaks) ▪️ Utility usage optimization ▪️ Dynamic rent pricing models The result? ➡️ Better decision-making. ➡️ Faster deal flow. ➡️ Higher NOI. But here’s the truth… AI is leveling the playing field. Information asymmetry is dead. The best CRE players won’t just know the numbers—they’ll know how to act on them faster than everyone else. AI won’t replace you. But the investor/broker/operator who uses AI better than you will. The future of CRE isn’t AI vs. Humans. It’s AI + Humans. The only question is—are you adapting? What AI tools are you experimenting with in your deal-making or operations? 📩 DM me or drop your thoughts below. ➖➖➖➖➖➖➖➖➖➖➖➖ "𝐓𝐡𝐢𝐧𝐤 𝐁𝐢𝐠, 𝐀𝐜𝐭 𝐒𝐦𝐚𝐫𝐭, 𝐈𝐧𝐯𝐞𝐬𝐭 𝐁𝐨𝐥𝐝𝐥𝐲." ➡️ I’m Logan Freeman, the #KansasCity #CRE Guy. 👉🏽 I can help you sell, buy, or invest in CRE in KC. 🫱🏾🫲🏼 Let’s talk, meet, and figure out how I or my team can help. #AI #CRE #Underwriting #RealEstateInvesting #CommercialRealEstate #Proptech #CRETech #DataDrivenInvesting

Chat GPT5 just dropped, so I asked AI to break down a public net lease property listing. 19 minutes later → it gave me something that used to take a half day to create. No client data. No lease files. Just publicly available info + a hypothetical rent roll. Here’s what it delivered: • Tenant profile → based on public credit ratings & filings • Market direct comps → from recent sales pulled off public listing sites • Cap rate sensitivity → at multiple purchase prices & exit assumptions • Lease abstract → compared to market norms for this tenant type • Risk assessment → investment red flags with supporting public data • Investor-ready summary → formatted so you could drop it into a deck today It wasn’t “AI assist.” It was a full professional-grade underwriting snapshot — in minutes. Why it matters: The workflows we thought were fixed… aren’t. In 3 years, the way we underwrite net lease deals today will feel as outdated as faxing offers. The speed of market analysis is about to change: • Brokers → Can price more deals per week • Investors → Can screen twice as many opportunities without hiring • Lenders → Can pre-screen risk with clarity Early adopters in CRE aren’t just saving time — they’re multiplying capacity. If you want to see exactly how I generated this, drop the word “snapshot” in the comments and I’ll send you the exact redacted prompt I used. #NetLease #CRE #CommercialRealEstate #RealEstateInvesting #PropTech #AI #ArtificialIntelligence #Underwriting #CapRates #InvestmentStrategy #DealFlow #RealEstateFinance #FutureOfWork #DataDriven

Yesterday OpenAI announced a ton of new updates and features to #ChatGPT and their #API. Here is my take on how these updates impact #AI in #realestate. 1.     Knowledge through April 2023 GPT-4 Turbo (used in ChatGPT Plus) now has knowledge of world events up to April 2023. That means it’s aware of the rise in interest rates and falling commercial real estate property values.  https://lnkd.in/enez8qTJ 2.     Ingests ~16x More Text GPT-4 Turbo has a context window of 128k tokens which is the equivalent of a 300-page book and 16x more than the previous model. This means larger documents like PCAs, environmental reports, JVAs, PSAs, #leases, OMs, #appraisals, etc. can be read and analyzed without specialized programming. This makes building an automated due diligence process much simpler (although humans should remain in the loop!). It can also write longer documents: research reports, investment memos, investor updates, appraisals, LOIs, etc. 3.     Cheaper! GPT-4 Turbo is multiples cheaper than the previous model. Again, this makes an automated due diligence process more realistic as it is far less expensive to process large documents. 4.     Enhanced Function Calling or One Window to the World Function calling lets you to build your own AI apps using external APIs. This opens up the possibility of having a single window into multiple APIs or SaaS applications (do we really need 15 tabs open for all our software). Function calling isn’t new, but they’ve improved the accuracy of the model. I will be demonstrating an example of Function Calling at the National Council of Real Estate Investment Fiduciaries (NCREIF) conference next week. 5.     Code Interpreter via API If you haven’t used Code Interpreter in ChatGPT, it can ingest data in various formats and perform mathematical operations as well as generate graphs and tables. Code Interpreter via API means the pipes now exist to build parts of the quantitative underwriting and memo generation process (e.g., sales/rent comp selection, tenant concentrations, lease rollover schedules, covenant calculations, stress testing, impact on existing portfolios, etc.). 6.     AI with Eyes The API now accepts images as input. Let’s send it all the images taken during property tours and inspections. What physical issues will AI be able to identify better than a human? Roof issues? Mold? Prior floods? What else? 7.     Text-to-Speech I look forward to AI summarizing property updates and reports and reading them to me in a natural sounding voice.  Scroll down on this page to listen to OpenAI voice samples. https://lnkd.in/eU-gBPDm Like and connect for more in AI in real estate! 👍 #cre #commercialrealestate #privateequity #brokerage #leasing #appraisers #fintech #proptech #investments #investmentmanagement #assetmanagement #acquisitions #cmbs #mortgageservicing

As corporate workplace and real estate professionals consider the office component of their portfolios, how will AI agents impact them? What will this do to the demand for office space? Currently the hot topic in the AI saga, we know that these agents have many characteristics of real people. AI agents can make decisions independently and even partner with human colleagues to together tackle many complex and varied tasks. And this goes well beyond the existing capabilities in LLMs like ChatGPT. In many ways, the AI agents will become coworkers on the other end of digital interactions. This human-AI co-worker relationship will likely evolve such that AI coworkers can give their human colleagues recommendations, suggesting tasks and even guiding them through them. While this will not happen immediately, both human and AI coworkers may “manage” one another, depending on the context, in ways analogous to current human teaming models. Most organizations didn't bother to modify their business process and workflows as distributed work expanded beginning in 2020. The way we worked (at least on paper) stayed pretty much the same, regardless of whether it was done in an office or remotely. This cannot continue in the era of AI-agents. Managers will need to decide who will execute the work, humans, AI agents or both. Processes will need to be totally redesigned including the digital workflows that support them. And in the face of digital business process and workflows, likely new AI agents will monitor the huge new volumes of data coming from the digital work processes. This data will support the creation of automated continuous improvement, a long-cherished goal that is rarely achieved. Continuous improvement in the quality of the work, where work is performed, and whether humans or AI agents should execute it will be a driver of change. No doubt, AI agents will also impact office space in terms of the size of the footprint and the location of the offices. Just another reason for organizations to shorten their occupancy commitments and remain agile for the changes to come for almost every organization. And remember, AI agents require zero space....except on a server somewhere. Logic suggests that the more AI agents in place, the lower the number of employees needed for the same level of production. And yes, this likely means less office space overall.

So, let's talk about artificial intelligence and cities. 🛠️ We’re heading into a world where AI won't just assist decisions—it will make them and implement them. The New York Times journalist Cade Metz recently published a sharp article on the AI Futures Project, a think tank effort predicting how #AI will reshape life over the next two years. Here are my hottakes—and how they apply to the #realestate industry: 🔥 AI will change how we frame problems: Mastering supply chains and financing has always been the crux of success. Artificial Narrow Intelligence (ANI) helped us collect options and data. Generative AI (GAI) moves us into complex decision-making augmentation. With Artificial General Intelligence (AGI) on the horizon, the real shift will be rethinking what we build based on live supply chain, labor, and capital markets data, not static pro formas. 🔥 AI will shift which resources matter: Traditionally, the firms that could deploy capital expenditures strategically won in real estate. Now it’ll be the firms that can capitalize on data and decision loops, using information faster, smarter, and better than anyone else. 🔥 AI could accelerate the gap between bold and cautious players: Incremental adopters may survive, but the bold early movers will set the new standard for cost, speed, and resilience. In an industry built on margins, that's existential, though the reality is the speed of this change will be modulated by regulation and the very incremental and multi-party nature of how decisions are made in making buildings. From my experience as an architect and developer, I can see firsthand how badly we need incremental improvements to lead to change in housing affordability, project delivery, and capital flows. Today, we’re still fighting slow approvals, outdated financing models, and inefficiencies baked into design and construction. But the opportunity is massive. Check out the article here: NYT: https://lnkd.in/eYaQxSkh Tell me what you think below 👇 #Innovation PS. I am starting to read and think more about AI and how it will impact cities. Please share any interesting resources or articles you come across. 🙏🏽