Lawyers may see themselves or others in the profession as luddites – but, specifically in the context of construction law, we should reflect on the nascent integration of construction and artificial intelligence and prepare for what is to come.
The growing role of AI in the construction industry
In 2024, capital expenditures for non-residential construction were $230 billion for Canada, and $112 billion for Ontario. As of September 2024, construction made up $164.7 billion of Canada’s total GDP – approximately 7.3%. Construction is an important industry, both domestically and abroad, and despite notable moments (COVID-19 effects, recent economic downturns) has maintained a steady growth.
AI is a term to describe a machine mimicing human thought – pattern recognition, problem-solving, and learning. Machine learning is the use and development of computer systems that are able to learn and adapt without following explicit instructions, by using algorithms and statistical models to analyze and draw inferences from patterns in data. In recent years, there has also been a boom in generative AI – the use of generative models to produce text, images, videos, or other forms of data.
This article will not delve into the technical aspects of AI, whatever the terminology, AI is being developed, tested and used across clients’ businesses in various ways:
-
bidding phase: AI-assisted promote efficiency, and have the potential to create better bids for all stakeholders by predicting realistic cost, time and scope solutions; design phase: to suggest preliminary concepts, including by building 3D models via generative AI accounting for engineering, structural, mechanical, etc.;
-
financing stage: to assess profitability and risk, and to assist in predicting cost overruns using historical data being input into predictive models;
-
operations, procurement and asset management: to maximize utility and minimize costs by using common data environments and digital construction data, as well as the ability to automatically assign priority to tasks, to deliver resources, from training to materials and equipment, when it is needed where it is needed;
-
during construction
-
to track, via sensors, the progress of the work – for example, to assess in real time the dying of asphalt;
-
to autonomously control cameras (mobile or stationary) to take pictures of worksites to evaluate progress, or compliance with safety protocols;
-
autonomous assembly of various components of the building off-site cheaper and quicker, to be fitted together on-site;
-
using machine learning, tracking and evaluating progress in a grading plan to identify scheduling risks early, or summarizing and alerting to critical issues from a review of the documentation exchanged; and,
-
post-construction: to identify issues in the operation and performance of buildings, bridges and roads and integrating with service and maintenance teams earlier than currently possible.
Everything from watches and glasses, mobile devices, drones, security and proprietary sensors, and even our appliances can become a source of information to be fed as data, creating unique challenges. As clients increasingly incorporate AI into their projects, the legal landscape is also evolving. Design disputes, contractual ambiguities, and regulatory gray areas are just the beginning. Staying ahead of these changes will be essential for lawyers practicing in construction law.
For example, imagine an AI-generated design that results in structural defects due to flawed algorithms, incomplete data, or forgetting to account for local building codes. Who is liable? The software developer? The architect who trusted the system’s output? The contractor who executed the design? What was the reasonableness of AI-generated designs and related schedules? Does reliance on algorithmic recommendations satisfy the standard of care? What is the impact on sub-contractors downstream, and is any compensation owed to them as a result? Did the parties contemplate, and select, the right insurance? What is the proper avenue and timeline for the various claims? Should the contract drafters, or their representatives, have anticipated errors? What would be the recourse for lenders providing construction loans on the basis of AI work product? Lawyers will need to be prepared to discuss with their clients the inherent risks and opportunities as they continue developing.
AI adoption in construction law firms: Opportunities and obstacles
Within law offices, some potential uses are:
-
administrative: assisting with drafting proposals, budgeting and billing;
-
fact investigation: to summarize and review client agreements and documents when issues arise, for example for scope issues or the application of a force majeure clause; and,
-
drafting: from correspondence, to client agreements, to termination notices, to litigation materials
However, AI has not gotten much traction yet.
In part, this is because AI is dependent on data – this comes into direct conflict with the profession’s confidentiality obligations. The more general (and older) the data – the less useful the results will be. That is why an answer from ChatGPT will not be as precise as if a lawyer uploaded his entire client file. A lawyer would likely not want to upload all of their client’s materials – such as CCDC agreements, supplementary conditions, change orders and directives, e-mail exchanges, and other documentation – just to assist with making a decision tree, for example when determining the right to terminate for unpaid invoices.
The continuum runs from publicly accessible data to expensive proprietary systems that may take long to adequately train and will still not have access to data from other firms. Privacy and utility must be balanced – the financial considerations of proprietary software (training the machine in-house, and providing it access to self-update using the latest law) may pose too great a barrier of entry for all but the largest of firms.
Accuracy is another issue - by now, we are familiar with the concern of the Courts and law societies across North America as to the unbridled use of AI, such as relying on cases provided by ChatGPT in a factum. Law societies, including those of Alberta, Manitoba, Saskatchewan, British Columbia and Ontario, have prepared white papers about generative AI, such as using the technology for basic legal research, marketing, editing, summarizing and drafting documents. This is also likely to be a concern for legal insurers (such as LawPRO and excess insurers) into the future, if AI work is being integrated into a lawyer’s work product, as constant vigilance and quality control is expected and required. A lawyer that does not properly and carefully check everything prepared by AI risks making mistakes that could be costly for their clients and themselves, and this is not likely to change even when the technology matures.
In addition, there is an increased push to disclose to clients when a lawyer has used generative AI – which, a lawyer, justifying hourly rates that his client likely already considers high, may not wish to do to maintain a positive solicitor-client relationship.
Preparing construction lawyers for the future
Moving forward, construction lawyers should:
-
Enhance technical expertise: Learn the basics of the available methods and processes.
-
Consider revising contracts: Develop templates that address AI-related risks, such as liability for algorithm errors or data misuse.
-
Engage technical experts: Build relationships with AI specialists who can assist in interpreting complex systems before and during disputes.
-
Stay Informed: Monitor developments in AI regulations, AI solutions and emerging case law.
-
Form AI practice groups: Consider establishing dedicated teams within your firm to focus on AI-related issues.
Construction lawyers will not soon be replaced by AI, and the profession is unlikely to adopt any new technology quickly (barring Zoom during the pandemic), but with time, tools will come out that will better correspond the needs of lawyers, For now, we need to prepare for the future – both to advise our clients and to manage our own practices – in a way that incorporates the new tools, including by sharing information across legal industries to design best practices.
***
Gleb Matushansky focuses on commercial and construction litigation, involving a variety of disputes including breach of contract, oppression remedies, professional negligence, liens, deficiency and delay claims, and commercial and residential real estate conflicts. With a background at small and mid-size Toronto firms, he’s known for providing practical advice and passionate advocacy to swiftly resolve issues for his clients.
Active in the legal community, Gleb serves as a repeat mooting judge, offering valuable feedback to national undergraduate competitors. His dedication to mentoring and coaching aspiring lawyers showcases his commitment to the field beyond his own practice.