How to Use Artificial Intelligence to Help Solve the Growing Housing Crisis

Welcome back Jason Cassidy, the CEO of Shinydocs, an artificial-intelligence-based document management platform that automates the process of finding, identifying and using collective knowledge within a business data environment. Jason’s topic in this posting is about applying artificial intelligence (AI) to help solve a growing problem in Canada, and elsewhere, as rising interest rates, prices, and demand have inflated the cost of owning and renting a home.

Mainstream media has discovered AI in the last two years with the emergence of OpenAI’s ChatGPT, creating debates over the technology’s promise and peril. From Forbes Magazine to conversations around the water cooler, debates over AI’s potential impact on academia, supply-chain management, communications, and more have become commonplace.

Thought leaders, academics, and pundits have begun applying AI to social and political issues to test its value with the hope of solving longstanding challenges. One of those is home ownership affordability which has become particularly acute in Canada.

Recently an article in Macleans Magazine, a Canadian news publication similar to Times Magazine and The Economist, even predicted “the end of homeownership.” A recent CBC report (the CBC is Canada’s public national broadcaster) stated that “Canada’s rental crisis is getting worse, according to a new report that found the average asking price for rent in September was $2,149, up by more than 11% compared with a year ago.” Meanwhile, a Royal Bank of Canada report added d to concerns noting that in the third quarter of this year, households in Canada required 62.7% of their take-home pay on average to cover the cost of ownership thanks to skyrocketing housing prices, inflation and interest rates.

Canada isn’t alone in facing a housing affordability crisis. The United States has the same problem. In both countries, it is characterized by a growing gap between cost and income, and driven by:

1. Rapidly Rising Housing Prices: The cost of housing, particularly in major cities, has skyrocketed in recent years attributed to low supply, higher demand, and speculation.
2. Stagnant Wages: Income has not kept pace with rising costs making affordability an issue.
3. Mortgage Interest Rates: Before the COVID-19 pandemic interest rates had remained low for more than a decade. But inflationary pressures have caused central banks to raise lending rates which has caused the cost of mortgages to climb.
4. Low Housing Supply: The housing supply is not keeping up with demand. There is general agreement that the shortage is attributed to the lack of purpose-built rentals, restrictive regulations on new builds, long timelines for approvals, and rising land costs.

Canada’s various levels of government have struggled to solve the housing affordability issue. They have created programs, policies and strategies to stimulate the number of new builds, and have provided financial assistance programs for low-income earners, rent control measures, and changes to land-use zoning.

What they haven’t done yet is look at how AI could help. A computer scientist named Alan Kay once said, “Technology is anything that wasn’t around when you were born.” Applying AI to solving the homeownership crisis may in the future, therefore, seem as normal to homeowners as doorbell cameras, smart thermostats, rooftop solar, and battery storage systems are today. The question is not whether AI will affect the housing issue, but how?  

Rather than directly increasing the housing supply or focusing on boosting economic factors and incentives, what an AI solution brings to the issue is the ability to find efficiencies. The process of building new homes involves developers, inspectors, and government regulators. It generates tons of paperwork including applications, plans and design approvals. This is where AI can speed up our human-made systems to shorten the timelines so that shovels get into the ground faster. An AI can analyze the data contained in applications, plans, regulations, building codes and more to find faults or shortcomings in projects.

The catch, however, is an AI can only help if the quality of the data it sees is well-vetted. AIs are not just software algorithms or reinvented mathematically-grounded constructs and systems, but rather built from neural networks that mimic our brains and voraciously consume large datasets. When presented with raw or unstructured data, an AI neural network can over a short period, begin to see patterns and report human errors or inconsistencies. It can also propose solutions to the errors or inconsistencies it finds. This is how it can shorten the process of getting shovels into the ground so that more housing is built. 

From the invention of the telescope to the steam engine, humans have shown a willingness to embrace new technologies to make discoveries and solve problems. If housing data made available to an AI is properly vetted, then applying this technology to the current crisis could be a real game changer.

It was Yogi Berra who purportedly said, “The future ain’t what it used to be.” We cannot afford the future to be like the present or the past when it comes to solving the housing crisis. With AI helping we can ensure that solutions to availability and affordability can be found and implemented.