Why AI matters at this scale

Skidmore, Owings & Merrill (SOM) is a global architecture, engineering, and urban planning firm renowned for iconic skyscrapers and complex urban developments. With a workforce of 1,001-5,000 and nearly nine decades of history, SOM manages massive, multi-year projects involving thousands of stakeholders, intricate supply chains, and relentless pressure to innovate in sustainability and performance. At this enterprise scale, even marginal efficiency gains in design iteration, risk prediction, or resource optimization translate to millions in saved costs and enhanced project viability. The architecture, engineering, and construction (AEC) industry is historically fragmented and slow to digitize, but AI presents a paradigm shift. For a leader like SOM, leveraging AI is not just about keeping pace; it's about defining the future of the built environment, delivering smarter, greener, and more resilient cities faster and more reliably than ever before.

Concrete AI Opportunities with ROI Framing

1. Generative Design for High-Performance Buildings: The conceptual design phase is both highly creative and intensely analytical. Generative AI models can ingest site data, zoning codes, sustainability targets (like LEED or WELL), and programmatic needs to produce thousands of viable design options in hours. This compresses a weeks-long exploratory process, allowing designers to rapidly converge on optimal solutions for energy efficiency, structural integrity, and occupant comfort. The ROI is clear: superior building performance locked in at the earliest stage reduces costly engineering rework later and can command premium leasing rates, while the time savings allow architects to engage in more strategic, value-added client collaboration.

2. Intelligent BIM and Construction Coordination: Building Information Modeling (BIM) is central to modern AEC, but coordinating complex 3D models across disciplines (architecture, structure, MEP) is error-prone. AI-powered agents can continuously audit BIM models for clashes, code violations, and constructability issues, flagging problems in real-time. This moves quality assurance from a periodic, manual review to a continuous, automated process. The direct ROI manifests in drastically reduced rework and change orders during construction, which are primary sources of budget overruns and delays on large-scale projects.

3. Predictive Project Intelligence: SOM's vast archive of historical project data is an untapped asset. Machine learning can analyze past projects to identify patterns leading to budget overruns, schedule slippage, or specific material supply chain disruptions. By building predictive models, project managers can receive early warnings and deploy mitigation strategies proactively. The ROI here is risk mitigation—protecting profit margins and client relationships by avoiding the severe financial and reputational damage of a failed project.

Deployment Risks Specific to This Size Band

For a firm of SOM's size and global footprint, AI deployment faces unique challenges. Integration Complexity is paramount: any AI solution must seamlessly interface with a sprawling, established tech stack (e.g., Autodesk Revit, Rhino, enterprise ERP) across multiple offices without disrupting live projects. Data Governance becomes a major hurdle, as valuable project data is often siloed within regional offices or individual project teams, requiring significant effort to centralize and standardize for AI training. Change Management at this scale is arduous; shifting the workflow of hundreds of seasoned architects and engineers from traditional methods requires careful, phased training and clear demonstration of value to overcome institutional inertia. Finally, there is Professional Liability Risk; the firm must establish rigorous validation protocols for AI-generated designs to ensure they meet all safety, regulatory, and ethical standards, as ultimate responsibility remains with the licensed professionals.