Curiosity in the Age of AI

Artificial Intelligence (AI) will never be your most powerful tool for real estate appraisal.

 

With all of the rapid development of AI, you may think that assertion is bold. Maybe you have just begun experimenting with it and are amazed. Maybe you are well beyond experimentation and have extensive AI workflows, or maybe even near-autonomous AI assistants! Regardless of where you fall on the AI learning curve—and as useful as it may be—it will never become your most powerful tool. That is because any appraiser’s most powerful tool isn’t something external. Your most powerful tool is internal… curiosity.

Cultivating curiosity is one of the most important and most impactful things we can strive for. Now, to be fair, I think that is true whether you are an appraiser, accountant, carpenter, or barista at the local coffee shop.  

Prerequisites for Curiosity

Curiosity deepens our engagement with the world around us. It requires something of us. You can’t be passively curious. To be curious, we must notice, and then we must inquire.  

The act of noticing requires the active engagement of our senses. I’m reminded of this every day. While north Texas lacks natural beauty in many ways, the sunset is often something to behold. Many days, I’ll find myself driving west at sunset. My eyes are open, but I’m not actually noticing anything. My senses are passively engaged to keep me safe on the road, but it is not until I actively focus on the sunset that I’m able to see its beauty.

The act of inquiry comes from the desire to understand, to create or discern order from that which our senses bring to us. Humans are naturally inquisitive when we stop to notice.  

From an interpersonal perspective, curiosity is critical to forming and maintaining healthy relationships. But it is no less critical in understanding and predicting market behavior, which is the very foundation of the appraisal profession.

 

Analytical Curiosity

Now, we could spend some time applying this concept into several aspects of the valuation process, but I’m going to focus our attention on data analysis.

For analysis to be meaningful, it has to take into consideration the “shape” of our market. By that I don’t mean whether or not the market is “doing well”, “in equilibrium”, or a “buyer’s market”. As I’m using this term, the shape of the market is the collective interaction of how people are behaving relative to value-influencing features. The only way that we can begin to understand the shape of our market is to be curious:

• What property characteristics are driving demand?
• What features are common in this market segment?
• How does the market react to properties that differ from the norm?
• What locations might be considered by the typical buyer for the subject?
• If I were a buyer in this market segment, would I expect a discount for this quirky feature?

These are the type of questions that will begin to develop a conception of how our market functions, and even help us to begin forming a strategy to answer those questions. What we are looking for in understanding the shape of a market is causation. Why do people behave the way that they do?

When we look at the data, we need to be aware that data itself comes to us with its own shape. No set of sales data will ever perfectly reflect any given market, and even if it could, no formula, algorithm, or model will ever capture that perfection. Data alone can reveal correlation, but never causation, which is why we have to be curious about our data and the techniques we use to analyze it.

 

Curiosity and Technology

And this brings me back to AI. Noticing and inquiring can’t be outsourced or delegated.

Let me give you an example. I have a fun custom GPT that I built to analyze land sale data. When I upload land sale data, it will produce four different regression models along with land value predictions for each model. It also produces a graph for each one so I can have a visual representation. Sometimes I’ve found it helpful, but without curiosity, it is worthless… even dangerous. I have to notice and inquire throughout the process. How did I select the comparables? Are there variables which drive values that aren’t considered? How do those four models even work, and are they even relevant? Is my data set large enough to be meaningful?

Now, in that example, the GPT is not technically doing anything “artificial”. It is essentially just running a Python script and isn’t actually making any choices. When AI is actually generative, the importance of curiosity goes up exponentially. The questions must go a layer deeper. How can I best interpret the output? What assumptions were made? What other ways might I approach the same problem? How can I reconcile differing results from differing techniques? These are all questions—among many others—that need to be posed regarding any analytical tool (whether AI or not). AI may be able to discern a shape of the data, but has no way to extend that to the shape of the market which the data is supposed to represent. 

In the case of the custom GPT I mentioned, the only reason I use it at all is because I built it in by selecting mathematical techniques which match the shape of how the market behaves. I selected models that mimic the Law of Diminishing Marginal Utility, which is a common market behavior related to surplus land. So I don’t even consider using it when the problem involves excess land. I considered the shape of the market first, and then selected a data analysis technique which respects that shape.  That is always the order of things. Notice the shape of your market, notice the composition of your data, and then inquire as to the best ways to analyze that data in light of what was noticed.

 

Cultivating Curiosity

So how can we cultivate curiosity? The best way to cultivate curiosity sounds simple: being present. This may sound silly, but in our current technological age, we are all conditioned to have a short, fragmented attention span. Being present—your mind and body occupying the same physical, mental, and emotional space at any given moment—is a challenge.  Truly being present requires not just attention to your current time and space, but sustained attention.

Sustained attention is the requirement for inquiry to begin to form. The enemy of sustained attention is multitasking. The human brain cannot multitask. What we call multitasking is actually just rapid shifts in attention. Our current technological age has taken multitasking from something which seems like a necessity for efficiency (although studies suggest that it doesn’t actually succeed in increasing efficiency) to something which seems like leisure activity: scrolling. If you are scrolling on a phone, it may seem like you are sustaining your attention on the screen, but in reality you are shifting your attention from one snippet of content to the next with every swipe of your thumb.

Cultivate curiosity by training yourself to have sustained attention on one thing at a time. Not only will you find that inquiry will follow, but the shallow inquiry of “what” will begin to deepen into the “why” questions which help us to actually understand and discern meaning.

Cultivating curiosity with data means taking the time to actually consider the data you have found.  

One practical way of doing this is through data visualization. Histograms can be very helpful. I regularly look at histograms to help me get a sense of where the subject fits into the wider set of data. Histograms of gross living area, lot size, age, and various features help me to understand how the subject fits within the wider market. Noticing how the subject fits allows me to begin to inquire how the market might perceive the subject property.

AI and other analytical tools can be very helpful in creating histograms, but for those to have usefulness and meaning, we must take the time to be curious about what they might be communicating.

It should be no surprise to us that data analysis does not truly have an “easy” button, because nothing in life does. Just remember that just because something isn’t easy doesn’t mean that it must be drudgery. The things in life which require the most effort are often those things which provide us with the most satisfaction.