April 18, 2025

Can We Rely on AI for High-Stakes Decision-Making?

The integration of Artificial Intelligence (AI) into the construction and real estate sectors has introduced a paradigm shift in project management, cost estimation, and risk assessment. However, the "black box" nature of AI, its inability to explain its decision-making process poses challenges to accountability, transparency, and trust.

This blog post examines the implications of AI-driven decision-making in high-stakes projects, the ethical concerns surrounding its implementation, and the necessary frameworks for ensuring reliability and professional responsibility. The analysis further explores how AI is currently utilised within the construction and real estate sectors, the perspectives of industry professionals, and the long-term impact AI could have on industry standards and regulatory practices.

AI is increasingly being used to enhance efficiency and accuracy in real estate development, construction project management, and cost consultancy. Algorithms can predict project delays, optimise material procurement, and refine cost estimations based on large datasets.

However, the trust gap between AI-generated insights and professional decision-making remains a pressing issue. Professionals question whether AI's opaque processes can be sufficiently transparent, accountable, and ethical for use in high-stakes decisions, as highlighted by recent guidance from the Royal Institution of Chartered Surveyors (RICS, 2025).

In addition to regulatory considerations, there is an ongoing debate about the ethical implications of AI in high-stakes decision-making. Can AI truly replace human expertise in critical project evaluations, or does it serve as an augmentation tool that still requires professional judgment? Understanding the limitations and strengths of AI applications is crucial for the construction and real estate sectors to fully harness its potential while mitigating risks.

The "Black Box" Problem in AI Decision-Making

AI systems, particularly those powered by deep learning, often operate without transparent processes, leading to what's commonly referred to as the "black box" dilemma. This opacity complicates auditing and validation of AI-driven decisions. Assigning responsibility when AI systems provide flawed predictions such as inaccurate cost forecasts or underestimated project risks remains problematic.

Determining liability among developers, operators, or professionals becomes challenging. Moreover, professionals require clear explanations to trust and justify AI-driven decisions. The inability of AI to articulate the rationale behind its conclusions undermines its credibility, leading to skepticism and hesitancy in its adoption.

Furthermore, AI systems trained on flawed, biased, or incomplete datasets may yield misleading or skewed outcomes, reinforcing incorrect decision-making and enabling systemic biases. Legal practices for AI-driven errors remain limited, and ethical guidelines and regulatory frameworks are still evolving to adequately manage these complexities.

AI in Construction and Real Estate: Benefits vs. Risks

AI presents numerous benefits in construction and real estate, including enhanced predictive analytics capabilities that significantly improve forecasting accuracy for costs, timelines, and risk management. It also streamlines routine tasks such as document reviews, compliance checks, and procurement, thereby enhancing overall efficiency.

Additionally, AI facilitates dynamic cost management by enabling real-time budgetary adjustments based on evolving market conditions. Resource optimisation through efficient allocation of workforce and equipment reduces project delays and minimises resource wastage, creating significant cost savings.

In cost modelling, AI-driven forecasting tools are improving the accuracy of project budgets, helping to reduce unnecessary contingencies and align financial planning more closely with real-world market conditions. By quickly analysing large historical datasets, AI can identify potential scope gaps and validate contractor pricing, speeding up project approvals and reducing delays.

On the contrary, significant risks exist, notably the over-reliance on AI predictions which can inadvertently reduce professional oversight. Such dependence could result in errors going unnoticed, potentially leading to considerable losses. AI models, trained on limited or historical datasets, may also struggle with generalisation, failing to adequately handle complex or unique project scenarios.

As AI tools become more widely used, concerns are rising around data confidentiality. Many construction and real estate projects involve strict IP and confidentiality agreements. Using public, internet-connected AI platforms risks exposing sensitive project information. Organisations must ensure any AI use complies with contractual obligations and protects client data from inadvertent breaches.

‍

Perspectives from Industry Professionals

Surveyed industry professionals, including Quantity Surveyors (QSs) and Project Managers (PMs), express cautious optimism regarding AI's capabilities. While many recognise AI's potential to significantly enhance operational efficiency, they remain hesitant about fully trusting algorithmic predictions without professional validation. A fundamental view among these professionals is in favour of a hybrid approach, integrating AI as a complementary tool to human expertise rather than as a direct replacement. This perspective underscores the importance of maintaining professional judgment and expertise alongside technological advancements.

Within contract administration, AI is being trialled to support the review process by detecting conflicts, missing scope, and inconsistencies faster than manual methods. While AI can help streamline contract drafting and reduce time and costs, industry professionals emphasise that final risk evaluation still requires experienced human review to protect client interests.

Ensuring Trust in AI-Driven Decision-Making

To effectively address concerns around trust and reliability, it’s essential to prioritise developing explainable AI (XAI) models that can clearly articulate the reasoning behind their recommendations and decisions in a way that humans can easily understand. Clearly positioning AI as a supportive technology ensures that critical decisions remain under human professional judgment, thereby maintaining accountability.

Organisations must ensure any AI use complies with contractual obligations and protects client data from inadvertent breaches. Regular monitoring and auditing of AI models are essential to prevent errors arising from biased, incomplete, or outdated datasets. Adherence to regulatory compliance and ethical standards, such as those established by RICS (2025), is imperative to ensure responsible AI use.

Furthermore, comprehensive AI literacy programmes for industry professionals are necessary to facilitate a better understanding, interpretation, and critical evaluation of AI-generated outcomes. Equipped with such knowledge, professionals can effectively leverage AI while preserving the necessary human oversight.

Variation forecasting is another emerging application, where AI analyses historical project patterns to identify potential variation risks early. Early detection enables better cost control and more proactive budgeting — crucial in managing the risk of budget overruns, which remain a major cause of financial disputes in the construction sector.

Looking ahead, several key developments will shape the responsible adoption of AI in high-stakes decision-making. Research advancements in explainable AI models promise greater transparency, potentially sustaining trust and facilitating broader acceptance by industry professionals. It is also anticipated that stronger governance and regulatory standards will emerge from both industry and governmental bodies, further solidifying AI accountability frameworks. As AI tools mature and their reliability improves, industry-wide adoption is likely to become increasingly prevalent.

Additionally, organisations may establish AI ethics boards dedicated to overseeing AI-driven decisions and ensuring ethical compliance, transparency, and professional integrity. These oversight bodies will become crucial in managing AI integration responsibly, ensuring alignment with established professional standards and ethical norms.

AI holds transformative potential for the construction and real estate sectors; however, fully realising this potential requires overcoming significant trust barriers associated with AI-driven decisions. Through the implementation of explainable AI technologies, robust regulatory frameworks, continuous auditing, and comprehensive professional training, the industry can responsibly integrate AI into critical decision-making processes.

Ultimately, AI’s role in high-stakes environments must remain focused on enhancing, not replacing human judgment, ensuring a collaborative future where both technology and professional expertise coexist harmoniously.

References

Royal Institution of Chartered Surveyors (RICS, 2025). Professional Conduct and Responsible Use of AI.
Boston Consulting Group (BCG, 2025). AI in Construction: Enhancing Efficiency While Mitigating Risk.
Emirates News Agency (WAM, 2025). AI Growth in the UAE: Trends and Developments.
McKinsey & Company. (2024). The Future of AI in Real Estate and Infrastructure Management.
Harvard Business Review. (2024). The Explainability Dilemma in AI Decision-Making.
World Economic Forum (WEF). (2025). AI Ethics and Responsible Automation in Professional Services.

April 18, 2025

Can We Rely on AI for High-Stakes Decision-Making?

Latest News