Briefing Note #2
Intro to the Briefing Note
The purpose of this note is to help individuals and organisations in the real estate world navigate AI and its potential impact on the industry.
Everything will be grounded in the question “and what does this mean for the real estate industry”.
This briefing note will provide you with insights, news, and tools in a format that you can read or listen to every fortnight. We will look to lessons from other industries and progress in our industry. Questions that you should be asking yourself and your company in your work.
The ultimate purpose is to leave you feeling more confident in navigating the world of AI in Real Estate.
This note is produced by Hattie Walker-Arnott, who trained as a chartered surveyor, worked in commercial investment & development, and then moved to a tech company. She has focussed on AI since mid-2022 alongside working as a development manager for regeneration projects.
5 practical steps for investors and developers (pt 1/5): Start trialling Microsoft Copilot Pro
Why? Early tests found it saves people on average 14 minutes a day. Not groundbreaking. But it adds up to around 4-5 hours a month ⁓£6 per hour saved. In theory, the more you use it, the more useful it is, the more time you (and your team) save.
Initial Uses - It’s pretty magical. It’s like having a ChatGPT trained on your own files. You can ask it “what options did we present for that project on XYZ street a few years ago?” and it will tell you. Importantly, it cites its sources so you can check the information is from the right place.
It’s embedded into most Microsoft applications. It’s useful in Word & Powerpoint. Outlook is potentially great but is currently buggy. It doesn’t have a wide range of uses in excel yet.
Did you know you had Digital Debt? Microsoft is framing Copilot Pro as a solution to “Digital Debt”. Microsoft says “the inflow of data, emails, meetings, and notifications has outpaced humans’ ability to process it all.” (link).
My cynical view is that they are framing it this way to dodge the “AI is taking jobs” attacks.
Either way, the real estate industry is an email-heavy, document-heavy industry. Real estate professionals have a huge amounts of inbound information coupled with numerous daily meetings and site inspections. In my experience in this industry it’s normal to take a week+ to reply to an email. When I worked in tech, they couldn’t fathom why it would take more than a day or two to respond.
Anything that can alleviate the load is positive, in my book.
Security & access - security is a key concern for any AI product. Microsoft Copilot Pro seems to have resilient security barriers. It only has access to the files that you can access, saved in OneDrive or Sharepoint.
A major barrier is that many organisations don’t have rigorous OneDrive/Sharepoint saving policies. Copilot Pro is more useful when it has access to more files.
Just remember if you don’t have access controls properly set up, any team member might be able to search with Copilot Pro “bonuses 2024” or “redundancy list Q3 203” and find things they are not meant to find.
Proceed with caution, but I would still proceed.
Cost (if you’re a Microsoft user, which in my experience most real estate people are). It costs around £360 per person for a year (as a business you have to buy a whole year’s subscription at once).
Glossary
LLM - Large Language Model
A Large Language Model (LLM) is an artificial intelligence model designed to understand, generate, and manipulate human language. Trained on vast amounts of text data, LLMs utilize deep learning techniques to perform tasks such as language translation, text summarisation, question answering, and code generation. Note GPTs are a subset of LLMs (all GPTs are LLMs, not all LLMs are GPTs). They were never officially “coined” in the way GPTs were in scientific papers but have emerged in language in the early 2020s.
Key Features:
Natural Language Processing (NLP): LLMs excel at interpreting and generating human language.
Contextual Understanding: They can grasp context, enabling coherent and relevant responses.
Versatility: Used in applications ranging from chatbots and virtual assistants to automated content creation and software development
Quote of the Fortnight:
“Technology’s unavoidable challenge is that its makers quickly lose control over the path their inventions take once introduced to the world.”
Thought of the Fortnight
Will software as we know it exist in 10 years time? - Software exists to provide non-technical users with easy-to-use tools to achieve their goals with technology. With AI now capable of generating code from natural language, will we simply request the necessary code for our desired actions, eliminating the need for traditional software products?
Key News Stories of the Fortnight
Microsoft’s 2023 Emissions are 20% up on 2022 (link) – Microsoft’s annual Environmental Sustainability Report shows 2023 scope 1,2 & 3 emissions are up 20% since 2022. They are 30% up since 2020 when they announced their 2030 plan to become carbon-negative, water positive, zero waste, and protect more land than they use. This brings into question how realistic their ambitions are. Their report mentioned AI (125 times…) as a potential solution: AI could improve measurement, increase data centre efficiency and improve energy transmission. New data centres are purpose-built for the power demands of AI and no water is used in their cooling. Putting the solution on AI feels like a bit of a cop-out. Achieving its sustainability goals by 2030 feels ambitious especially when companies are increasing their compute in order to stay competitive (see the Musk’s new fundraise below…)
Musk raises $6bn for xAI (link) - xAI, Musk’s AI venture started in mid-2023. This fundraise values it at $26bn. They state “xAI is primarily focused on the development of advanced AI systems that are truthful, competent, and maximally beneficial for all of humanity. The company’s mission is to understand the true nature of the universe.” Funds will be used to take xAI’s first products to market, build advanced infrastructure, and accelerate R&D. Through Musk’s other companies X, Tesla and Neuralink, xAI has access to incomprehensible volumes of written, visual and biological data, which no doubt will be used to train xAI. According to The Information, Musk wants to build a Supercomputer 4x bigger than any currently in existence (link).
Self-driving vehicles due on British roads by 2026 as the UK’s Automated Vehicles Act became law (link) Human error occurs in 88% of accidents. Therefore the government is hoping that deaths and injuries from drink driving, speeding, tiredness and inattention could be drastically reduced. From a real estate perspective, autonomous vehicles could increase the value of less accessible places. Additionally, they could change how we design car parking, imagine if your car could take itself off to a car park 5-minutes away having dropped you where you need to be. Transport planners will need to get their heads around the potential of this pretty fast. The government estimates the AV industry could be worth £42 billion and create 38,000 more skilled jobs by 2035. This news also comes as London-based autonomous vehicle start-up Wayve raised $1billion, the UK’s biggest AI fundraise according to TechCrunch.
Note: following the general election announcement no party AI strategies have been released yet. We’ll update once they have.
AI Fundamentals - Artificial Neural Network
Exploring frequently used terms
Artificial Neural Networks are fundamental to the development we’ve seen in AI in recent years. However, they spent decades out of favour with computer scientists. It was only in 2012, when computing power became strong enough to process vast amounts of data that they came to public attention.
They power the facial recognition in our phones and traditional voice recognition like Siri and Alexa. Now they form the architecture of LLMs like Chat GPT.
Definition: an artificial neural network is a computational model inspired by the human brain, consisting of interconnected nodes (neurons) organized in layers.
Purpose: Artificial neural networks are designed to recognize patterns, learn from data, and make decisions or predictions.
How? Artificial neural networks work by simulating the behavior of the brain's neurons. They consist of interconnected nodes, organized in layers. Each node receives data as an input, processes it and passes the result to the next layer. Connections between nodes are given different weights, effectively different levels of importance, that adjust during training.
Artificial neural networks are mostly trained by a method call backpropagation. When the network makes a mistake, backpropagation works out how much each connection contributed to the error. It then adjusts the weights of these connections to reduce the error. By repeating this process, the network learns and gets better at making accurate predictions.
Disclaimer: This content is for informational and educational purposes only. It does not constitute financial, investment, legal or IT advice.