Introduction and Overview
We were delighted to present this collaborative workshop organised by the Society of Property Researchers (SPR), Cambridge Finance, and McDermott Will & Emery. The event, titled ‘AI for Real Estate: Practical Applications,’ aimed to explore the transformative impact of artificial intelligence on the real estate sector. The workshop focused on advancements in forecasting, valuation, and property research, with practical demonstrations on how AI tools can enhance productivity and streamline analytical processes for research, valuation and legal.
This hybrid event was held in-person and virtually on 4 February 2025, from 08:30 to 11:00 GMT, at McDermott Will & Emery UK LLP, 22 Bishopsgate, London, EC2N 4BQ. The event was supported by Real Estate Women.
Speakers
- Maria Wiedner, CEO, Cambridge Finance
- Dr. Nikodem Szumilo, Associate Professor, The Bartlett Faculty of the Built Environment, UCL
- Hunter Jackson, Chief Knowledge Officer, McDermott Will & Emery
- Rosemary Tulloch, Senior Associate, McDermott Will & Emery
The seminar opened with a contextual overview of AI, tracing its evolution from the 1950s to the present. Dr Nikodem Szumilo noted that while the underlying mathematics of AI, particularly trial-and-error learning models, has been known for decades, the convergence of computational power, data abundance, and pre-trained models has propelled AI into mainstream use.
The recent enthusiasm surrounding AI stems from these pre-trained models, like ChatGPT, Gemini, and DeepSeek, which allow users to access advanced capabilities without the need for extensive computing power or training data. The discussion highlighted generative AI, which can produce novel outputs resembling creativity, though the nature of AI-generated creativity remains a philosophical debate.
Session 1: Analysis, Strategy, and Forecasting with Generative AI
Speaker: Dr. Nikodem Szumilo, Associate Professor, The Bartlett Faculty of the Built Environment, UCL. Dr. Szumilo is a leading expert in urban economics and real estate finance. His session explored how Generative AI (GenAI) can integrate quantitative and qualitative data into strategic real estate analysis. Key topics included the use of AI for forecasting and scenario planning, and a step-by-step approach to strategy formulation using AI, covering:
1. Gemini 1.5 (Deep Research)
- A Google-powered research tool capable of conducting large-scale internet searches across hundreds of websites within minutes.
- Produces comprehensive, referenced reports with citations.
- Users can edit research plans to refine focus areas, such as narrowing a search to specific geographic regions like Canary Wharf offices.
2. ChatGPT Operator / Float Equivalent
- Designed to automate tasks on a user’s computer, such as setting up email accounts.
- Currently more of a conceptual demonstration with limited real-world effectiveness.
3. Kimi
- A Chinese AI model that performs rapid searches and summaries.
- Provides impressive functionality but raises privacy concerns as it collects user data, including keystrokes.
- The open-source nature of Kimi allows users to download and run the model locally for enhanced data security.
4. Google AI Livestream
- Offers live screen-sharing with AI guidance.
- Assists with tasks like Excel operations in real-time, providing step-by-step instructions.
Demonstrations and Applications
Research Assistance with Gemini 1.5
A demonstration showed Gemini 1.5 conducting research on European office markets. The tool generated a structured research plan, reviewed 64 sources, and produced a detailed, referenced report within minutes. This illustrated the speed and efficiency AI brings to research-intensive tasks.
Interactive Support Using Google AI Livestream
The speaker shared their screen with Google AI, requesting help with creating a pivot table in Excel. The AI provided real-time guidance, though technical issues with Excel limited the demonstration. The exercise highlighted AI’s potential for on-the-job learning and support.
Task Scheduling with ChatGPT
The speaker demonstrated ChatGPT’s task-scheduling feature, enabling sequential task automation, such as:
- Researching office market trends in London.
- Conducting deeper analysis on two identified trends.
- Compiling a final report. Though limitations exist due to task size restrictions, this feature showcases AI’s ability to automate multi-step processes.
Reasoning Models: DeepSeek
DeepSeek was praised for its capacity to display its reasoning process. When queried about interest rates’ impact on the housing market, DeepSeek outlined its thought process step-by-step before delivering an answer. This transparency helps users understand the logic behind AI-generated outputs, making it particularly valuable for teaching and research.
Session 2: Practical Applications of Generative AI with Excel and Python
Speaker: Maria Wiedner, CEO, Cambridge Finance. Maria Wiedner, a specialist in applying GenAI for financial modelling and data analysis, led a practical session demonstrating the integration of AI with Excel and Python. She showcased how AI can:
- Automate data extraction and manipulation
- Streamline reporting processes
- Enhance workflow efficiency This session provided valuable insights into combining traditional Excel-based workflows with AI-powered coding and data analysis.
Real Estate Financial Modelling Use Case
- Excel Retention: Due to data privacy concerns, Maria Wiedner emphasised retaining Excel as the primary tool while leveraging AI for coding assistance and automation.
- Python and Excel Integration: She illustrated using Python libraries (e.g., Pandas and ExcelWings) to process large datasets, such as VOA rateable values.
- Mapping and Geolocation: By integrating Python with Excel, she automated the creation of maps displaying property comparables based on geographic data.
- Prompting: Maria Wiedner stressed the importance of clear prompts to guide AI effectively, such as specifying the extraction of address data from unstructured VOA datasets.
Session 3: Legal GenAI Overview + UK Real Estate Industry Speaker: Hunter Jackson, Chief Knowledge Officer at McDermott Will & Emery. Hunter Jackson is the Chief Knowledge Officer at McDermott Will & Emery, driving the firm’s innovation in knowledge management and generative AI. He co-leads the deployment of cutting-edge GenAI solutions like ChatGPT and CoCounsel, while overseeing a department that is building solutions which capture, organise, and share the firm’s collective experience, research, and precedent.
Together with Rosemary Tulloch, Senior Associate at MWE, this session explored the potential of generative AI for the UK real estate industry, focusing on tools like Thomson Reuters’ CoCounsel. Hunter Jackson provided a live demonstration of CoCounsel and offered an overview of how McDermott is developing legal AI use cases and solutions, with practical examples tailored to the UK real estate market.
Legal Practice Application
Rosemary Tulloch and Hunter Jackson demonstrated CoCounsel, an AI platform designed for legal document analysis:
- Document Review: Capable of processing complex leases and development agreements (500+ pages).
- Question-Answer Extraction: Generates concise answers to legal queries, with citations linked to specific clauses within the document.
- Verification Necessity: Despite its capabilities, human oversight remains essential. Issues such as handwritten amendments and ambiguous clauses can lead to inaccuracies.
Key Takeaways
- Efficiency Gains: AI drastically reduces the time required for research, data extraction, and document review.
- Complementary Role: AI enhances, but does not replace, human expertise. Users must verify AI outputs.
- Data Privacy: Secure platforms and local AI installations are critical in legal and real estate sectors handling sensitive data.
- Tool-Specific Strengths: Different AI models excel in distinct tasks. Users should match tools to specific use cases.
Future Implications
The session concluded by addressing the future trajectory of AI in real estate and legal fields:
- Electronic leases and integrated property data systems could streamline workflows further.
- Drafting capabilities in AI platforms are improving but are not yet fully reliable.
- Legal professionals must adapt to AI or risk being outpaced by tech-savvy peers.
The session underscored that AI adoption is a journey requiring experimentation, verification, and a balance between technological potential and professional judgment.
Q&A Session: Tailored Solutions for Your Use Cases
The seminar ended with participants engaging in an interactive Q&A session, presenting their own real estate data challenges and receiving expert advice from Dr. Szumilo, Maria Wiedner, Hunter Jackson and Rosemary Tulloch on how AI can enhance productivity and solve practical problems. Some key questions included:
- How secure is it to upload client data into AI platforms like ChatGPT or Gemini?
- Can AI detect discrepancies or conflicting clauses within a legal document?
- Is it possible to automate running multiple variations of a query to generate diverse results using AI?
- How does AI handle handwritten or manually amended legal documents?
- Are there any tools that can help predict legal outcomes, such as winning an argument in court, based on precedent and data?
- Will AI eventually replace traditional legal drafting, or are we moving towards standardised electronic leases?
- How can AI integrate with existing property management and valuation systems for better data flow?
- What are the practical steps to combine Python scripting and Excel for property data analysis without compromising data security?
Conclusion
The workshop demonstrated the growing significance of AI in transforming real estate research, financial modelling, and legal practice. AI tools are enhancing efficiency, automating complex processes, and providing innovative solutions to longstanding industry challenges. As these technologies continue to advance, professionals in the real estate and legal sectors must remain informed and adapt to leverage their full potential.
If you are interested in joining the Society of Property Researchers (SPR), Cambridge Finance, Real Estate Women (RE:Women), or would like to learn more about legal advice and AI solutions at McDermott Will & Emery, please feel free to contact us via email at info@cambridgerefinance.com and we will direct you accordingly.
Photo Gallery
