David Pfeffer, Casey J. Rahn and Sean Scuderi | Tarter Krinsky & Drogin
The integration and utilization of artificial intelligence (“AI”) into our lives is no longer a distant or speculative concept reserved for dystopian fiction; it is already embedded, and only becoming more deeply entrenched, in the daily operations of modern businesses, including within the design and construction industries. Architects, contractors, and developers are increasingly integrating AI platforms into workflows to enhance efficiency, generate design options, and streamline analyses. As design and construction professionals alike continue to embrace these tools more widely, exploring both the transformative benefits and associated implications of their use can facilitate more informed and effective implementation.
AI in Construction: Liability, Accountability, and Oversight
When AI platforms or machine learning tools are used as the sole basis for assessing design adequacy and structural integrity, the resulting risks extend beyond theoretical concerns but may potentially compromise the physical integrity of constructed facilities themselves and significantly increase liability exposure for all relevant stakeholders. Aside from the obvious constructability issues of the project itself, the delegation of diligence to automated systems raises significant concerns regarding user accountability, particularly where confidential or proprietary information is involved. To address these risks, stakeholders should adopt disciplined and contractually-defined protocols for AI utilization within their project agreements, no matter where a party sits in the project hierarchy.
Seven Tips for Mitigating AI-Related Risks in Construction
- Agreements should clearly define the permissible use of AI tools. This includes, at minimum, requiring advance notice to contracting parties concerning how a party intends to utilize AI platforms as part of their scope of work; the particular platform(s) of intended or authorized use, including AI platform version identification; the party’s intended scope of use; and which deliverables may incorporate AI-generated content.
- Should AI platforms be authorized for use on a particular project, the agreement should mandate independent human oversight provisions for all AI-generated outputs. This includes representations that licensed, qualified professionals will review, verify, and validate AI outputs prior to presentation and reliance. Human reviewers must be tasked with identifying inaccuracies, inconsistencies, or so-called “hallucinations,” and ensure that all work product adheres to the applicable standards of care. AI may assist a party’s evaluation, but it cannot replace professional judgment.
- Parties should implement robust confidentiality and data security provisions in any agreement governing a party’s AI use. Parties should establish clear protocols governing the handling of proprietary and confidential information, including restrictions on disclosure and use. Indemnification clauses should address intellectual property risks, particularly where AI-generated content may inadvertently infringe upon third-party rights. Contracts should also define ownership of both AI inputs and outputs, supported by appropriate user warranties regarding lawful use and authorization.
- Minimum security requirements for AI platform usage must be established. This includes compliance with industry cybersecurity standards, encryption requirements, and access controls designed to mitigate unauthorized use or data breaches.
- Ownership and usage rights concerning the original contract documents and AI-generated materials must be clearly delineated. Without explicit terms governing title to the work, disputes may arise regarding (i) whether AI outputs are deemed derivative works for purposes of party ownership; (ii) who retains rights to underlying data; and (iii) how such materials may be reused or shared. Restrictions on downstream use can help preserve ownership rights and limit unintended exposure. Any party utilizing AI software must also verse themselves in the AI-platform’s user policies to determine the platform’s ownership parameters over any data inputs/outputs.
- Given the evolving nature of AI platforms, agreements should integrate ongoing transparency obligations regarding software updates, version control, and model performance. Parties should be notified of material changes to AI tools, including updates that may impact outputs or introduce model drift. Maintaining records of AI inputs, outputs, and decision-making processes is crucial for traceability, dispute resolution, and regulatory compliance. Additionally, parties should include incident response protocols addressing the identification, reporting, and mitigation of data breaches or system failures.
- Insurance requirements should be revisited to account for AI-related risks as traditional professional liability policies may not fully address exposure arising from AI use. Parties should evaluate whether specialized coverage or endorsements are necessary to adequately protect against claims involving design errors, data breaches, or intellectual property disputes linked to AI systems.
Final Thoughts
While AI offers transformative potential for the design and construction industry, its use in determining design sufficiency demands careful governance. By proactively addressing contractual, technical, and risk management considerations, stakeholders can harness the benefits of AI, while safeguarding against its inherent uncertainties.
When one of your cases is in need of a construction expert, estimates, insurance appraisal or umpire services in defect or insurance disputes – please call Advise & Consult, Inc. at 801.641.8304, or email experts@adviseandconsult.net.