Starting with:

Step 1: Develop AI Fluency (Not Just Literacy)

AI literacy is knowing what AI is. AI Fluency is knowing how to use, critique and apply it.

Literacy can get as complicated as you like, but I think, as business people, there are just a few things that are essential. What is Predictive AI, Generative AI and Causal AI? And what is AI Bias, and what are hallucinations? And the difference between automation and augmentation.

So:

Predictive AI is a branch of artificial intelligence that analyses historical data, identifies patterns, and uses machine learning algorithms to forecast future outcomes, enabling proactive decision-making across industries. Fundamentally this is an analytical tool.

Generative AI is a type of artificial intelligence that creates new content—such as text, images, audio, and code—by learning patterns from existing data and generating novel outputs that mimic human-like creativity and reasoning. Fundamentally this is a creative tool.  Causal AI is a branch of artificial intelligence that goes beyond correlation-based predictions by understanding cause-and-effect relationships, enabling more explainable, reliable, and intervention-driven decision-making. (This is a nascent field but will become important. For obvious reasons)

AI bias refers to systematic errors or unfair outcomes in artificial intelligence systems that arise from biased data, flawed algorithms, or human-driven assumptions, often leading to discriminatory or unbalanced decision-making in areas such as hiring, lending, and law enforcement. We joke about CRE being ‘pale, male and stale’ - which is a prime source of AI bias to look out for!

AI hallucinations refer to instances where an artificial intelligence system generates false, misleading, or nonsensical information that appears plausible but is not grounded in reality or factual data, often due to limitations in training data or model reasoning. There are increasingly good ways to mitigate this, but as with humans ‘Trust, but verify’.  

The above alone, well internalised, is a good start but for AI Fluency it is different. Predictive and Causal AI are highly technical, complicated fields that frankly you are not going to be fluent in without a lot of study. Generative AI, however, is just a matter of practice. It does help to have some training in good prompting, but if you just interact with a frontier LLM as you would with a human, you’ll be well on your way. Just keep asking questions. After about ten hours of use you will have a very good feel for what works or doesn’t, and their strengths and weaknesses. Each model has a distinct ‘character’ and particular strengths: Claude excels at coding and natural writing, ChatGPT is strong for business-related writing and strategic thinking, while Gemini’s vast working memory makes it ideal for analysing large documents.

The key though is practice. You need those 10 hours under your belt. Incorporate asking a model about everything you do, and you’ll soon get there.

PS. For imagery, cough up $10 a month (at least once) to play around with MidJourney - it’s the best.

Step 2: Reframe Your Mindset: AI as a Co-Pilot, Not a Replacement

Make no mistake - AI is going to take a lot of jobs. Simply put, AI will dramatically improve productivity in many areas, and so unless the market for these areas grows significantly we will need less people to generate the outputs required. We need a bigger pie. I am confident this will emerge in many sections of real estate but certainly not in all. So we need to ensure we are in the growth areas.  

And this is going to be where humans + AI can either do X better, faster, cheaper or where humans + AI enable things to be done that were hitherto impossible. This latter state is known as achieving AI Synergy. When combined the end result is better than either the best human, or the best AI, can achieve on its own.  

An example, from Portfolio Optimisation: An AI might propose optimisation strategies (e.g., leasing adjustments or refurbishment needs), but human expertise is essential to evaluate feasibility and, crucially, align these with investor expectations. This AI Synergy enables a more data-informed, holistic outcome.  

YOU have to be working where human + AI is required, but don’t kid yourself that is everywhere. Even if that’s true today, you need to be looking a year or two ahead. ‘Is what I am doing something an AI is likely to be able to do ... sometime soon?’

As you get more familiar using the tools this will become clear. You will ‘feel’ what is going to be possible.

Step 3: Focus on the Uniquely Human Skills

AI can handle pattern recognition, automation, and prediction. Anything ‘structured, repeatable, predictable’ will be eaten by AI.

AI can also simulate creativity, judgment, empathy, and complex problem-solving, but it IS simulating these things. And in certain situations one doesn’t need more than ‘simulated’. But you need to discern where first-principles thinking, nuanced decision-making, emotional intelligence, and adaptability is the real value-add, because these are human skills that AI cannot (at least yet) provide.  

Being a ‘high quality’ human is going to become a super skill. Sounds odd but some humans are more human than others. And this is going to matter.  

As is Critical Thinking - you need to be really good at this in a world where nothing, necessarily, is true. Take ‘The TDH Daily CRE Critical Thinking Challenge’.

Step 4: Become a Master of 'Prompt Engineering'

The ability to interact effectively with AI will become as fundamental as typing or using the internet. You need to learn how to craft effective prompts that yield precise, valuable, and creative outputs.

Good prompting makes a real difference. Being able to ask a great question matters.  

The joke is that LLMs are ‘the revenge of the Humanities graduate’ because all of a sudden understanding language is important, and the ability to use words clearly, to explain context, and to elucidate constraints, can be the difference between a genius or a trivial answer from an LLM.  

You can scout the internet for good guidance on prompting—or simply take my #GenerativeAIForRealEstatePeople course. The latter is easier:)

Step 5: Understand Data and Its Value

Real estate has historically been data-poor, but AI is going to change that. We’ll soon have access to unprecedented amounts of information, and this will be much more democratically spread around. The days of data asymmetry are likely coming to an end (for reasons we’ll cover in another newsletter).  Which is all well and good but not much use if you know nothing about data science and analytics. Knowing how to interpret, validate, and apply data is a skill you must learn.

My top tips relate to one man, Bill Schmarzo, the so-called ‘Dean of Big Data’. He’s a leading authority on data science, analytics, and business value creation through data, and is known for his practical, business-focused approach to leveraging data as an asset.

Top tips are: read his book ‘AI & Data Literacy: Empowering Citizens of Data Science’ and his blog here.

Honestly, unless you’re a data science specialist, this will cover everything you, as a business person, need to know about data.

Step 6: Adopt an Experimentation Mindset

AI is evolving so rapidly that you shouldn’t get bogged down in perfection. A lot of the best outcomes, today, will come from experimenting. Using new tools, pushing them hard, and seeing what comes out the other end. Once again, you’ll get a feel for what is likely to be possible 6, 12 months hence.  

Use AI as a sandbox for innovation—experiment with different use cases. Pro versions of ChatGPT, Claude and Gemini (which includes NotebookLM which we covered recently) are less than $60/£50 a month. Just as an individual this is money well spent. As a company leader, give these to as many of your employees as are interested. Let them experiment and watch what happens.

Step 7: Prepare for New Business Models and Paradigms

AI is or will be reshaping industries, and legacy business models are sure to be disrupted. Understanding quite how this will or might shift value creation is not easy, but I think there are some ‘directions of travel’ for the real estate industry that are clear. For example, AI will likely shift ‘knowledge work’ towards human-centric skills, leading to more hybrid and distributed working, but also more intensive, purposeful collaboration. ‘No one knows what happens next’ is a good mantra to follow, so adaptability and flexibility are going to be at a premium. And at a more fundamental level, the rise of AI is going to necessitate a mass of new energy infrastructure, and a heap of data centres.

I read a paper yesterday suggesting we might see ‘a century of change in a decade.’ That feels hyperbolic, but it’s certainly wise to anticipate significant transformation in the coming years.

Step 8: Think in Terms of ‘Space as a Service’

Which brings me back nicely to my hobby horse ‘#SpaceasaService’.

In the world of AI-optimised work and cities, physical spaces will need to become smarter, more adaptive, and experience-driven. AI will drive dynamic demand forecasting, flexible workspaces, and real-time decision-making in urban environments, and being on top of the intersection of AI, automation, and physical infrastructure will be a key competitive advantage.

But above all else AI will enable us to provide real #SpaceasaService - Spaces that provide the services that enable every individual to be as happy, healthy and productive as they can be.

AI is going to enable us to ‘Build a Better Built Environment’. This is surely the point of being in the real estate industry?

Step 9: Cultivate an Anti-Fragile Career and Business Approach

As we’ve discussed, AI will automate many traditional jobs, but new roles will emerge. So we need to design and build careers that thrive in volatility—where adaptability, interdisciplinary knowledge, and AI augmentation are core. This will require a portfolio of skills, projects, and networks that position us as ‘AI-powered professionals’. Following the other 9 steps will, I think, deal with this one!

Step 10: Think About AI Ethics, Governance, and Human Implications

I am full of positivity about the next decade but it could go very badly wrong. WE need to ensure the future of AI is not just about what it can do but also what it should do. Sometimes it feels very dull and dreary but we must engage with discussions around AI bias, transparency, and responsible usage.

AI is a ‘General-purpose technology’ and these ‘affect an entire economy. They have the potential to drastically alter societies through their impact on pre-existing economic and social structures.’  Dull or not, this is serious stuff!

Final Thought: Be Curious, Not Fearful

One needs to think of AI not as an existential threat (though it might be) but as a transformative enabler. A technology that ‘enables’ us to do the currently impossible. With it we should be able to deal with bigger challenges. And looking around, the world is full of them. There is no shortage of work to be done. Should we choose to do it.

The winners in an AI-mediated world will be those who embrace learning, adaptability, and have a deep curiosity, and dare I say it, love for the world around them.

Hopefully that’s you.

OVER TO YOU

Which of these 10 steps are you implementing first? Let me know. Need any help?