Andrew Woolsey is an associate and Sophie Ashcroft is a partner in the commercial dispute resolution team at law firm Browne Jacobson.
Construction companies are using artificial intelligence tools to increase productivity and add new skills in areas such as contract management and project management processes. However, the use of AI can also lead to complex, and sometimes unexpected, disputes.
As a general point, the vast majority of construction disputes are resolved out of court or through private arbitration proceedings in which the hearing and final award are detailed. they are confidential Instead of real-life examples of AI solutions working poorly in a construction context, we’re using theoretical scenarios to set out the potential disputes that could arise in an AI project.
AI complicates accountability
Construction projects are often multifaceted and customized arrangements that change as the project progresses and involve multiple stakeholders.
In the overall context of the project, different contractual relationships exist. Developers have contracts with contractors who in turn have contracts with subcontractors and architects. The use of AI introduces an additional party to the chain, namely the AI developer.
“The use of AI means that typical accountability frameworks may not be adequate”
The question of who is responsible when construction projects go wrong is ultimately a question of fact, but the introduction of AI makes answering that question much more difficult due to the inherent complexity of the tools AI, which can trigger a whole host of costly and time-consuming disputes. .
For example, was the project failure due to faulty data, inadequate training methodology, or hardware issues with the AI tool? Architects and contractors can be held responsible for oversight and implementation failures. In the event that a construction project runs into difficulties, there is likely to be a cascade of claims, with interested parties seeking to recover their losses from the next party in the contractual chain.
Claims typically arise from a breach of contract due to service delivery failures (eg, missing contractual milestones or the AI solution producing results that do not meet specifications). This can lead to claims for damages, contractual remedies such as late payments and, in extreme cases, termination of the contract.
Advice for contractors
It is essential that stakeholders are clear about the main goals of the project and the ways in which AI will be used to achieve those goals. Contracting parties should ensure that comprehensive agreements are in place at each stage of the contract chain to provide certainty if something goes wrong. The roles, responsibilities, expectations and risk of the different parties should also be contracted and defined in detail from the start of the project, in order to distribute responsibility along the chain as clearly as possible.
The use of AI means that typical accountability frameworks may not be adequate. Parties contracting for the use of an AI tool in a construction project should ensure that the relevant agreements include AI-specific warranties, indemnities and limitations. These terms should be tailored to the specific context in which the AI tool will be deployed and should be based on standards that are clearly measurable. This would likely involve drafting warranties that, while based on common service standards such as reasonable care and skillthey respond to the fact that an AI tool is being used.
Where AI outputs are subject to human supervision, the scope of these obligations should be clearly spelled out, including identification of the required skills/experience of the individuals concerned, the nature of any training required, as well as the processes to be followed in the tests. monitor and review AI results. Keeping records in relation to reviewing decisions made by AI solutions can also be useful in managing the risks associated with their use.
It is also important for the parties to recognize and mitigate the risk that the AI developer (often a start-up) may not have sufficient assets available or sufficient insurance coverage to satisfy a claim, which could result in increased liability for the other parts of the chain.
Risk management is crucial to avoid disputes. Therefore, contractors should take a proactive approach towards risk management by reviewing and amending standard agreements which, for the reasons specified above, are unlikely to be suitable for projects using AI solutions. Contractors should work with partners, consultants, attorneys (whether in-house or outside counsel) to produce agreements that provide contractual certainty in an area that is inherently uncertain due to the complexity of the technology involved.
