In my role, I take a lot of sales calls from companies that have real estate or workplace technology that can help enterprises. They range from simple dashboards to complex iWMS and everything in between. One consistent that I have noticed since ChatGPT became a household word is that they are all spending massive amounts of attention and development time on integrating AI into what they do. It is fascinating to hear the variety of responses they give when they are asked “why AI is the future of their product.”
For those not quite aware, there are a variety of types of Artificial Intelligence. ChatGPT is an AI tool built on a large language model that generates text in response to prompts from a user. Its responses are based on the vast amounts of information that it has been trained on. Midjourney is another well-known AI that generates images in response to prompts from a user. Collectively, these tools have given rise to the new profession of Prompt Engineering. This is a job where an individual determines the best prompts to feed into an AI to get a certain or optimal result.
Machine Learning is a related field where algorithms are given a set of rules, parameters, and desired outcomes to follow. The algorithms are then set free to learn for themselves within those guidelines. One of my favorite examples of effective Machine Learning was AlphaGo by DeepMind. This is the self-learning algorithm that ultimately learned to play the game Go and defeated the world champion. In some senses, it taught itself to play the game and learned strategies and techniques that no human had ever considered.
At their core, AI and ML tools take lots of historical data, create new data, learn from outcomes, and use it all to come up with better ways to do things. They are complex fields that are seeing growing numbers of consumer-grade options as well as open-source options that can be applied in new and novel ways.
This brings us back to AI and ML in real estate tools. Corporate Real Estate is not by its nature a world with lots of big data. Ours is a messy world with lots of disparate data points that may or may not accurately reflect what is actually happening. To get true big data would require technology that actively tracks everything happening in an office which, thankfully, is not currently something companies are considering. The challenge for any given CRE technology is that they may not have enough data on their own to use AI or ML in a way that actually gives good outcomes. Without the full picture of how space is being used, these tools are limited in what they can achieve.
But that is not stopping Workplace Technology companies from marching fearlessly toward an AI/ML future for their systems. The thought seems to be that the first to create the right platform could be the one to reap the rewards and become the system of choice by companies far and wide. They are not always looking to find the targeted minimum viable implementation that fits who they are with focused use cases.
Where AI/ML works best is when there is a strong use case backed up by historical outcomes, ongoing data generation, and an area of focus where ongoing value can be created from improved outcomes. This fits a lot of systems and areas, including real estate. But the crux of it comes down to figuring out the right use case first.