Special Topics Course 11.S969
11.S969 (H4) AI and Data Science in Real Estate
Faculty: Siqi Zheng
Time: H4 Tuesdays/Thursdays. 9:30-11: 00 am
Room: 9-354
Grade: Letter grade
Units: 6
Course Description:
The “AI & Data Science in Real Estate” course is designed to address the rapid transformation of the real estate industry by integrating artificial intelligence, data science, and strategic implementation. This curriculum covers foundational AI concepts and data analytics, practical applications in property valuation and investment analysis, resilience and risk management, and ethical considerations while emphasizing hands-on learning with industry-standard tools. By merging technical competency with business acumen, the course prepares students to tackle emerging challenges and drive innovation in real estate development, finance, and investment.
Innovation and practical application are woven throughout the program, featuring guest speakers from academics and practitioners, startups and major firms alongside immersive projects and case studies. Students engage with predictive modeling, and regulatory compliance while tackling challenges related to ethics, data bias, and changing management. Central to the course is the way AI enhances financial decision-making, using advanced analytics for valuations, risk assessments, and investment planning. Through projects and case studies, participants develop solutions that reflect real-world industry needs, illustrating how AI enables data-driven strategies that elevate financial performance and support organizational change in a dynamic urban and real estate market.
This in-person, 12-session, 6-credit course gives MSRED, DUSP, MCP, and Sloan students a rigorous yet applied foundation in AI methods and how they are transforming the real estate value chain, from data ingestion and modeling to product design and portfolio strategy. Students will learn AI and Data Science fundamentals, work with real data and tools, analyze live industry use cases, and develop an implementable AI solution proposal that integrates financial, cultural, environmental, and regulatory considerations. Throughout the course, guest speakers from PropTech, major owners, tech firms, and government complement faculty-led sessions from CRE, Sloan, DUSP, and CSAIL.
AI is becoming embedded in valuation, leasing, development, and asset management workflows, and early-career professionals who can translate between technical teams and investment/urban decision-makers will have a distinct advantage.
The rise of AI-driven data centers, smart buildings, digital twins, and alternative data is changing location decisions, risk assessment, and ESG strategy.
Cultural signals, bias, and regulation mean AI in real estate is not just a technical problem but a governance and ethics challenge that practitioners must be prepared to navigate.
12-session structure and content:
Session 1 – AI Foundations for Real Estate – James Scott
Outline of class
History of AI in Rel Estate and Different Applications
Session 2 – Data Foundations and Machine Learning Basics – Albert Saiz
Looking at big data types and sources:
Spatial data: types and data architecture
Spatial data: sources and best practices
Session 3 – Industry Landscape and Opportunity Mapping – James Scott
Mapping the AI in real estate ecosystem: PropTech segments, venture trends, platforms, and capabilities across brokerage, lending, asset/portfolio management, and planning.
Workshop: students identify a high-impact AI opportunity in their domain (developer, lender, city, operator) and draft a one-page business/problem statement to refine over the term.
Session 4 – AI-Powered Financial Analysis – Walter Torous
Using AI for underwriting: scenario generation, cash-flow forecasting, risk scoring, and portfolio analytics; integrating AI outputs into Excel-based or Python-based pro formas.
Session 5 – Decoding the Invisible: Culture, Preferences, and AI – Siqi Zheng
How AI can surface “soft” drivers of value such as aesthetics, cultural features, school quality, and amenity mix, and link them to price and absorption outcomes.
Role of generative and computer-vision models in quantifying curb appeal, Feng Shui-type features, and culturally specific preferences, alongside risks of bias and stereotyping.
Session 6 – The Physical Cloud: Data Centers, AI, and Location Strategy – Siqi Zheng
Economics and siting of AI/data center infrastructure: power and cooling constraints, grid capacity, land use, incentives, and community impacts.
Use of convolutional neural networks and satellite/remote sensing data to identify and monitor data center sites, and to quantify environmental and option value effects on land markets.
Session 7 – Advanced Methods I: CSAIL Perspectives
Guest lecture from CSAIL on current AI research relevant to the built environment, such as deep learning for imagery, reinforcement learning for control, or multimodal models for planning.
Session 8 – Advanced Methods II: Sloan Perspectives
Sloan faculty session on AI strategy and organizational change: where AI sits in corporate strategy, build–buy–partner decisions, and capability building for real estate firms.
Case discussion of a real estate or infrastructure company’s AI transformation, including ROI analysis, pilot design, and scaling challenges.
Session 9 – Location Intelligence, Alternative Data, and Market Analysis – Calandra Cruickshank
Guest lecture (e.g., StateBook) on combining traditional economic development data with alternative sources (mobility, firmographics, news, social, satellite) for site selection and policy.
Session 10 – Robots or Digital Twin – James Scott or Guest Lecture
Digital twins for buildings, campuses, and cities: fusing BIM, sensors, and simulation with AI for operations, resilience, and scenario analysis.
Session 11 – Ethics, Regulation, and Future Trends – Guest Lecture
Bias, fairness, explainability, and discrimination risks in lending, tenant screening, pricing, and neighborhood analysis; links to housing equity and displacement.
Regulatory landscape for AI and data in housing and real estate (fair housing, credit, privacy, emerging AI regulation) and implications for compliant product design.
Session 12 – Final Presentations
Student project presentations with feedback from faculty
Suggested final project concepts (Session 12)
Students work in small teams on a scoped, implementable proposal; they can be more strategic/blueprint or include lightweight technical prototypes:
AI-Enhanced Valuation and Cultural Signal Platform
Build a concept for a valuation tool that integrates traditional comparables with computer-vision and text-based signals (e.g., images, listings, neighborhood descriptions) to surface “invisible” cultural and aesthetic value drivers while explicitly managing bias and fairness.
Deliverables could include a data schema, model approach, UI wireframes, governance plan, and a brief financial/ROI case for a developer, lender, or appraisal firm.
Data Center / AI Infrastructure Siting and ESG Strategy Toolkit
Design an AI-supported toolkit that helps an operator, utility, or city evaluate and prioritize potential data center sites under power, environmental, regulatory, and community constraints.
The project might combine remote-sensing concepts, grid and zoning data, and a scoring model, plus an implementation roadmap and stakeholder engagement framework.
AI Copilot for Real Estate Investment or Urban Planning Teams
Propose a domain-specific AI assistant that ingests an organization’s leases, financial models, zoning codes, and planning documents to support underwriting, permitting, or community planning.
Students define the knowledge sources, workflows (e.g., lease abstraction, scenario Q&A, memo drafting), risk controls, and a phased deployment plan inside a real or hypothetical organization.
